School: Lafayette Tecumseh Jr High School
Grade: 8
Category: ENBM
Abstract: This student has not yet submitted an abstract.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Honey Creek Middle School
Grade: 6
Category: BEHA
Abstract: This experiment investigated what real time effects listening to different music genres has on the human brain, focusing on mindfulness, calmness, and stress reduction in school aged children. It was hypothesized that listening to a Sanskrit shloka would generate brain wave patterns associated with mindfulness similar to those observed during meditation, thereby reducing stress and anxiety. Participants listened to four auditory conditions—silence pop, classical music, and silence—for three minutes each, with one minute intervals between sessions. Brainwave activity (alpha, beta, gamma, delta and theta), which are associated with calmness and focus was measured using a SereniBrain headband. The instrument categoried brain activity as active, relaxed, or calm, with active states linked to higher stress and relaxed and calm states associated with increased alpha, theta, and sensorimonitor rhythm (SMR) brainwaves and lower stress. Music was delivered through Bose QuietComfort headphones. Results showed that listening to the Sanskrit shloka produced the greatest increase in brain waves associate with mindfulness, followed by classical music, pop music, and silence. These results suggest that incorporating music—especially Sanskrit shlokas—during the school day, particularly during high-stress periods such as midday, may help reduce stress and anxiety, while improving focus. Future studies can examine the effects of longer listing durations and different age groups.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Honey Creek Middle School
Grade: 7
Category: EAEV
Abstract: For my project, I decided to look at the severe weather in my area. I wanted to look into this because I have had a lot of storms that have taken the power out in my area and that have damaged a lot of trees. This made me wonder if there is a trend going on that has an increase of storms or if this is a couple of outlier events. I gathered my data on the storms from the NOAA severe storm database and then entered the data into an excel spreadsheet. After this, I analyzed it by running a two tailed t-test which is how I got my results.
Elevator Pitch Video:
School: Stonegate Elementary
Grade: 4
Category: EGSD
Abstract: I wanted to find out if water or wind makes more electricity. Fossil fuels like coal, oil, and natural gas make most of our electricity, but they also make carbon dioxide, which is bad for the Earth. Water and wind are clean, renewable energy sources that don't pollute. If we use more clean energy, we won't need as much fossil fuel. My question was: Will a water turbine or wind turbine make more electricity? My hypothesis was that the water turbine would make more electricity. I built a water turbine and a wind turbine. I used the same template and materials to build the turbines. They both had the same number of blades. I used a 15W water pump to make the water flow and a 15W fan to make the wind. Both ran at full speed. The turbines connected to generators using the same pulley system. The only thing I changed was the source—water or wind. I measured each generator’s voltage with a multimeter. My hypothesis was right! The water turbine made more electricity. Both turbines worked by capturing the kinetic energy from the moving water or air and spinning the generator to make electricity. Since water has more mass than air, it has more energy when it moves. The heavier water pushed the turbine blades to spin faster. When the generator spins faster, it makes more electricity. Water power makes more electricity than wind power when both use the same amount of power to start.
Elevator Pitch Video:
School: Marian High School
Grade: 10
Category: ENBM
Abstract: Current prosthetic arms can cost from 2000 to 5000 dollars just for an immovable cosmetic arm that has no replication of any motion present in regular arms joints. There are thousands of people including many veterans who have either lost or were born without an arm that can't afford one, even the cheapest option. The main focus is developing a joint, specifically an elbow joint, that can effectively replicate the movements of a regular elbow weather that be extension (straightening the arm) or flexion (bending of the arm).Also developing it in a 3d software then printing it to help with affordability. I started my building in Tinkercad where I developed my idea for an elbow joint based off of certain hinge mechanisms that can be found on knees joints. Once the joint was finished I fastened two wooden dowels on either end of the joint with some glue and tape and glued a mannequin hand on the end of the “forearm” dowel. For my testing I gathered all the degrees of extension and flexion in the elbow for three different age groups of male and female. I then took the measurements of the prosthetic and matched it up to one of the groups. The closest gender and age it resembled was an adult female. The project proved successful because I proved that a simpler more cost effective alternative can be developed for the expensive prosthetic joints on the market today.
Elevator Pitch Video:
School: Noblesville High School
Grade: 11
Category: CBIO
Abstract: Cancer remains a major global health burden, with ~20 million new cases and 9.7 million deaths in 2022. Yet it comprises over 200 molecularly distinct subtypes, and real-world datasets are heterogeneous and subject to domain shift. Within-cancer cross-validation is an incomplete proxy for clinical readiness, since models performing well in-distribution often degrade on unseen tumor types. We introduce OmniSurv, a transformer-based, foundation-style model that learns a reusable, cancer-agnostic patient representation from six modalities: whole-slide histopathology, RNA expression, copy-number variation, somatic mutations, clinical variables, and retrieval-augmented biological knowledge. OmniSurv integrates (i) norm-balanced semantic tokenization to map disparate modalities into a shared embedding without dominance, (ii) cancer-conditioned cross-attention with prototype-based conditioning that generalizes to unseen tumor types without labels at inference, and (iii) a genomic recovery and anti-collapse design to preserve weaker omics streams. Trained on 10 TCGA cancers (n=3,137), OmniSurv achieves validation C-index 0.78, exceeding recent multimodal pan-cancer models (~0.72–0.76). Under cancer-type-level zero-shot holdout across 22 unseen cancers (n=3,455), it attains C-index 0.63 with significant stratification (HR 2.01), outperforming clinical Cox baselines (~0.53) and prior multimodal methods evaluated under the same protocol. The learned embedding strongly supports multiple downstream tasks without retraining. Zero-shot cancer classification reaches 95.4% accuracy; patient retrieval exceeds 96% recall@10; biomarker prediction achieves AUROC 0.92 (KRAS) and 0.86 (TP53), Tumor Mutational Burden R² ~ 0.87; survival calibration time-dependent AUC 0.81; subtype discovery separates IDH1/2 glioma with 0.89 purity; out-of-distribution detection achieves AUROC 0.93, enabling safe fallback. OmniSurv establishes a single, transferable cancer representation that generalizes to unseen tumors and powers multiple clinical tasks, moving beyond one model per cancer prediction toward deployable precision oncology.
Elevator Pitch Video:
School: Culver Academies
Grade: 11
Abstract: Circulating tumor cells (CTCs) are a clinically valuable biomarker for cancer prognosis and treatment monitoring, yet their extreme rarity and heterogeneity make reliable isolation challenging. Current clinically approved platforms primarily rely on immunoaffinity capture targeting epithelial markers, which can fail to detect CTC subpopulations undergoing phenotypic transitions and often suffer from limited throughput and cell loss. To address these limitations, this project presents the design of a label-free dual-field microfluidic chip that integrates inertial microfluidics and acoustophoresis for high-efficiency, size- and property-based CTC separation. In the proposed system, an upstream inertial microfluidic module exploits size-dependent lift and Dean forces to pre-focus larger tumor cells and enrich them toward a designated streamline while removing smaller blood components. Downstream, a standing surface acoustic wave field is introduced to apply differential acoustic radiation forces, enabling fine discrimination between CTCs and leukocytes with overlapping size distributions. To ensure long-term operational stability, the chip incorporates a hybrid anti-adhesion strategy, combining microscale SiO₂ micropillar arrays to maintain elevated near-wall shear stress with antifouling surface chemistry to suppress nonspecific cell–surface interactions. Numerical modeling and benchtop microbead experiments are employed to guide channel geometry, flow conditions, and field parameters, establishing a systematic design framework for dual-field coupling. By unifying passive inertial separation, active acoustic manipulation, and anti-adhesion engineering, this work proposes a scalable and clinically adaptable microfluidic architecture for robust, label-free CTC isolation, with potential applications in precision oncology and liquid biopsy workflows.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Culver Academies
Grade: 12
Abstract: Atmospheric carbon dioxide (CO₂) emissions have been majorly influenced by cement production. Microbial-induced calcium carbonate precipitation (MICP) has emerged as a method for sequestering CO₂ in a stable form by utilizing calcium ions (Ca2+), the by-products of cement production. For this experiment, Bacillus subtilis was investigated, and it secretes both carbonic anhydrase and urease enzymes that accelerate calcium carbonate (CaCO₃) mineralization. To test the MICP efficiency of B. subtilis under different environments, four conditions – with urea, without urea, with urea & 50 mM HEPES buffer, and with urea & 100 mM HEPES buffer – were tested. The efficiency was measured by comparing the final CaCO₃ mass yield for each condition. The data show that MICP efficiency is optimal when both urea and HEPES buffer are present in the media. There was not a significant difference between the 50 mM and 100 mM HEPES buffer conditions.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Guidance Islamic School
Grade: 8
Category: ENEV
Abstract: Access to clean drinking water is a global issue, especially in areas without electricity or advanced water treatment systems. The purpose of this project was to test how different materials—glass, plastic, and metal—affect the effectiveness of a DIY solar-powered water purifier. This project is important because it investigates how material choice can improve a low-cost, environmentally friendly method of purifying water using renewable solar energy. Three identical solar-powered water purifiers were built using the same design and filtration layers of gravel, sand, and activated charcoal. The only difference between the purifiers was the outer container material: one was made of glass, one of plastic, and one of metal. Equal amounts of contaminated water were placed into each purifier and exposed to direct sunlight for six hours. Water quality was measured by recording clarity on a scale of 1–5, the amount of purified water collected in milliliters, and pH levels before and after purification. The results showed that the glass purifier produced the best overall results, collecting the highest amount of purified water and showing the greatest improvement in clarity and pH. The plastic purifier showed moderate improvement, while the metal purifier produced the least purified water. These findings suggest that materials that allow more sunlight to pass through improve the evaporation and condensation process. Overall, this project demonstrates that material selection plays an important role in the efficiency of solar-powered water purification systems.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Saint Joseph School
Grade: 4
Category: PHYS
Abstract: For the Cub Scout Pinewood Derby, I wanted to find out what shapes and weight placement would give me the fastest car. I believe that the fastest shape of toy vehicle would be thre car named "Streamline" and for weight placement, the car with the weight distributed in the back would be te fastest. I took three different cars with different shapes ("Bus", "Streamline", and "Griswold") and raced them down the track to determine which shape would be the fastest. I completed two different heats with three races a piece to allow each car to race within the three different lanes. These cars were different shapes, but I added weights to make sure they were the same weight. I then took three identitical cars and placed weights in the front, middle, and back. I completed two different heats with three races a piece to allow each car to race within the three different lanes. For my results, in the "Shapes" race, the overal winner was the car labeled "Streamline". For the "Weights" race, Car #1 (with the weights in the back), won every single race of the two heats. I found that my hyptothesis was correct for both the "Shapes" and "Weights" placement in the toy vehicles. When I help my dad cut my own Pinewood Derby car this year, we will make it short in the front and slowly increase in height and place the weights in the rear of the Pinewood Derby car.
Elevator Pitch Video:
School: West Lafayette Jr/Sr High School
Grade: 11
Category: BEHA
Abstract: The use of Generative AI or GenAI among high school students is rapidly increasing. Concerns about over-reliance and other potential negative impacts have raised questions about whether GenAI is helpful or harmful for students’ learning. Despite its widespread adoption among high schoolers, few studies have examined GenAI use among high schoolers. To address this gap, we surveyed 49 high school students to investigate the following research questions: (a) How do high school students use GenAI?, (b) How is GenAI use associated with students’ goal orientation?, and (c) How is GenAI use associated with learning behaviors and outcomes? By answering these questions, this study aims to provide insights to help students to use GenAI in an appropriate way and inform teachers to guide students’ GenAI use to support learning.
Elevator Pitch Video:
School: Zionsville Middle School
Grade: 6
Category: EAEV
Abstract: 2.1 Billion people globally do not have access to clean drinking water, that is 26% of the world population. This experiment was done to find effective ways to purify water with simple materials that are easily obtainable. In this experiment I created simple homemade filters using simple materials and quick setup. I used a spectrophotometer to visualize the contaminants in the water that would be harmful to drink. In total, the carbon filter had less contaminants in the water than the sand filter, and the absorbance followed a trend similar to high purity reverse osmosis water and ultra-pure lab-grade water.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Zionsville Middle School
Grade: 5
Category: PHYS
Abstract: To understand why some karate kicks are stronger, I researched how the leg acts like a lever and how kinetic energy works. I learned that speed is the most important factor when determining the strength of a kick because doubling the speed quadruples the energy. The purpose of my project was to find which kick is the most powerful so I can use that information to help me win karate tournaments. My hypothesis was that the rear roundhouse would be the strongest because it uses the most body rotation. To test this, I set up a camera on a tripod and used a tape measure to record how far a kicking bag moved when hit. I performed three trials of each leg for the rear snap, rear roundhouse, and side thrust kicks. The data showed that the rear roundhouse was the strongest, moving the bag an average of 23.17 inches. The side thrust kick averaged 21.42 inches, and the rear snap kick averaged 19.42 inches. These results supported my hypothesis. The roundhouse was the most powerful because it uses the "kinetic linking principle" to coordinate body segments and maximize force. Even though I am left-footed, I was surprised to find my right leg was stronger in two of the three kick types. If I were to do this experiment again, I would have more people from my dojo participate and include other types of kicks.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Frankfort Middle School
Grade: 5
Category: MATS
Abstract: The purpose of this project is to determine which produces the most sliding friction: aluminum foil, plastic wrap, felt, sandpaper, cotton fabric, or printer paper. I predicted sandpaper, cotton fabric, felt, and plastic wrap would produce the most friction because when you feel them, your hand is not sliding much over these surfaces. I think sandpaper will make the most friction because of its texture, which is rough and bumpy. In the same way, I predicted that felt, aluminum foil and paper printer will have the least friction. With those surfaces, your hand slides easily. Those surfaces are slick (if the aluminum foil is flat). The paper is the smoothest, so it should have the least friction. To do this experiment, I got a long wooden board, a protractor, and a heavy weight. For each material, I secured it on top of the board, using tape. Then I put the weight on top of the material. I slowly raised the board until the weight began to slide. When that happened, I stopped moving the board and measured its angle from the table with the protractor. I repeated this process so that each material was tested five times. My results show rougher surfaces produced the most friction, and smoother surfaces produced the least. Sandpaper had the highest friction, and it is the roughest surface I tested. Similarly, printer paper had the lowest friction, and it was the smoothest. In the case of sliding friction, the roughest item creates the most friction.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Frankfort Middle School
Grade: 8
Category: CHEM
Abstract: The purpose of this project is to see if changing the temperature of sodium hydroxide in a hydrochloric acid titration affects the amount needed to neutralize the titration. I believed the neutralization of the would happen with fewer drops when the sodium hydroxide is heated up because if the molecules are sped up by the added heat, any mixing will happen faster, which would also make the titration will happen faster. If the molecules are slower with colder sodium hydroxide, they are moving less, which slows down the mixing, so the titration will happen slower with more drops needed to make the neutralization evident. For the experiment, I added hydrochloric acid and phenolphthalein into a beaker. Then I warmed up/cooled down sodium hydroxide with a hotplate/freezer. I added sodium hydroxide one drop at a time, then swirled to mix the chemicals. I did this until the color neutralization occurs. I repeated the test three times for each of five temperatures. My results show that as the sodium hydroxide is heated up, the number of drops needed decreases. This proves my hypothesis to be correct. I think this happens because when liquids are heated, the molecules making up the substance move slightly faster. When they are cooled, the molecules slow down. This cooling results in the substances not mixing as readily as they do when they are warmer, which slightly delays the neutralization of the titration. Colder temperatures will not really affect the titration much, while heat will.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Frankfort Middle School
Grade: 8
Category: BEHA
Abstract: The purpose of this experiment is to determine if being bilingual improves how fast your brain can complete two competing memory tasks. I think bilingual students will be faster at successfully completing competing memory tasks (in this case, putting letters and numbers in alternating order) with fewer mistakes than non-bilingual students. Bilingual students have to switch between languages from home to school, and they also are involved in conversations with friends during the day that use both languages. Therefore, their brains are used to switching between two different things. Being bilingual myself, I understand how it feels having to switch between both languages, having done so since I was four. Most bilingual kids have done this process of juggling two languages for a long time (such as when they are translating for someone else), so their brains are already trained to switch between two tasks and keep both of them straight. Non-bilingual students do not have their same sets of experiences, so I believe their brains will have a harder time with the switching back and forth required in this test. This will lead them to make more mistakes and finish the task with a slower completion time. The main part of this experiment that took some time was making the computer program that will do the testing. The program starts by asking three Spanish questions that are basic but not easy. They are simple enough for a bilingual to understand but too difficult for a non-bilingual who might just guess. Students answer each question. Next, a pattern task is explained. This involves alternating letters and numbers (up to 9), like this: A 0 B 1 C 2…the screen shows the pattern the student has to duplicate. Then this is a brief READY...SET...GO prompt for students to begin typing the pattern. Once a student is finished, the program ends. In a file on the flash drive, it records the pattern they typed and how long it took them to type it. Then I would go back later and access the file to see results to determine mistakes in the pattern and the speed of typing the pattern for each person who participated in the experiment. My initial classification method grouped all of the students who scored a “3” on the question part as bilingual, while all of the students who scored a “0” were classified as non-bilingual. Anyone scoring a “1” or “2” was discarded from the data set. These results of these two extreme groups were compared. My results from that initial grouping show that the bilingual students are, on average, about 4 seconds faster to complete the task and do so with a smaller number of errors (by about half an error per person). Because so much of my data was discarded in my initial classification method, I also conducted a second analysis where I included the “1” group as non-bilingual. This would add more data to my non-bilingual group and could also possibly close the gap between the average scores of the two groups. However, it did not. Adding more data to the non-bilingual group still produced the same gaps in results. This gives me further confidence that the differences in the groups are meaningful. Bilingual students were faster than non-bilingual students with the pattern task, and they produced the pattern with fewer mistakes. This occurs because the bilingual brain is already used to juggling multiple tasks and switching back and forth, so their brains have developed over time to being more efficient with managing multiple tasks. It takes a lot to learn two languages and fully understand how to translate them. So, having a better ability to manage memory tasks is highly beneficial. Bilinguals naturally benefit from this based on their environment and upbringing. I believe my project shows that teaching all students dual languages from an early age would be good for overall brain development. There is a benefit to knowing more than one language just from a learning standpoint, but if there is also a benefit in brain development, that is even more of a reason to start teaching students multiple languages early in life. So starting young with learning more than one language gives your brain greater capacity to work flexibly.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Frankfort Middle School
Grade: 7
Category: MATH
Abstract: The purpose of this project is to see if altitude affects the aspects of athletic performance related to fatigue in the NBA. When you are higher in the Earth's atmosphere, the air is thinner and contains less oxygen. That means an athlete’s body has to work harder to maintain performance and get the oxygen it needs through its red blood cells. When you are tired, you make mistakes and have a harder time putting effort into the game. I analyzed results from every game when the Oklahoma City Thunder went to Salt Lake City and Denver. I chose Oklahoma City because it is geographically close to the other two cities, is in the same time zone, and does not require as much air travel (reducing any jet lag and travel fatigue). I chose Salt Lake City and Denver because they are the two highest elevation areas in the NBA. I took three stats from the games starting in 2008: shots attempted, turnovers, and free throw percentage. I decided on these three stats because I think they could all be impacted by fatigue. I found that playing in an elevated arena did produce more turnovers from the team than what was usual. It also led to more shots. However, free throw percentage was not affected. Overall, this project shows that altitude does affect athletic performance in some aspects for teams that are not used to it. That could be a significant advantage if you play at a high altitude frequently.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Frankfort Middle School
Grade: 7
Category: CHEM
Abstract: The purpose of this project is to determine how pressure affects the rate off a Briggs-Rauscher oscillating chemical reaction. I believed that added pressure would speed up the cycle because the pressure will push the chemicals closer together, speeding up the rate of the mixing process. For each test, I would add hydrogen peroxide to a test tube, then starch solution, and finally potassium iodate. That starts the reaction, so I would immediately screw on the cap on the test tube. I quickly added pressure to the test tube to the desired amount using the air compressor through a secure tube in the lid. I held the tube by clamping it with my fingers to sustain the pressure. I waited one cycle of the reaction while adding the pressure and then counted ten cycles from that point forward. I timed this with a stopwatch. I repeated the test multiple times for each pressure amount. My results show my hypothesis is correct. As pressure increases, the cycles’ rate changes and speeds up. Although, within the experiment, I believe that it will eventually stop speeding up when you add too much pressure. The reaction will eventually stop gaining in rate when you add to much pressure because of the principle of diminishing returns. This principle explains that the benefits gained from something will eventually have a smaller gain as more is invested in it. So, pressure in a chemical reaction matters. The greater the pressure, the more the reactants mix and interact.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: West Lafayette Jr/Sr High School
Grade: 10
Category: CBIO
Abstract: Cancer is a global health challenge and leads to 10 million deaths annually. Heat shock proteins (HSPs), especially HSP90α, are frequently overexpressed in cancer patients and can be used as indicators of lung cancer. However, limited studies have been done to comprehensively synthesize its accuracy in lung cancer diagnosis. The goal of this study was to perform a systematic review and bivariate meta-analysis to evaluate the diagnostic performance and clinical utility of HSP90α for the detection of lung cancer. A systematic literature search was conducted across PubMed and PROSPERO databases. Following PRISMA guidelines, 8 high-quality diagnostic datasets comprising 3,624 patients (2,080 cancer cases and 1,544 healthy controls) were included. Statistical analysis was performed in STATA 19.0 to calculate pooled sensitivity, specificity, and the Area Under the Receiver Operating Characteristic (AUROC) curve. The results showed a pooled sensitivity of 71% and an exceptional pooled specificity of 92%. The overall AUROC was 0.84, indicating excellent discriminatory power. The Diagnostic Odds Ratio (DOR) was 30, and the Positive Likelihood Ratio (LR+) of 9.3 demonstrated significant clinical power. While high heterogeneity was observed (I2=98%), meta-regression confirmed that sensitivity and specificity remained remarkably stable regardless of the varied clinical cut-off values (ranging from 49.8 to 114.8 ng/mL). Deeks’ Funnel Plot yielded a p-value of 0.102, indicating no significant publication bias. The results in this study show that HSP90α is a robust and reliable biomarker for lung cancer diagnosis.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Central Catholic Jr-Sr High School
Grade: 9
Category: ROBO
Abstract: This project uses a smart camera to provide tools for animal care through the mobile identification of pathogens and the detection of toxic plants. The objective is to develop AI models that can easily classify microbes in cultures and also assist animal owners in removing toxic plants harmful to animals and livestock. I collected images and trained the AI to identify common pathogens, such as Aspergillus and E. Coli, and toxic plants. I then put the AI models on a smart camera that has a microscope lens attachment. This could be an inexpensive education tool that can also be used for field applications.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: West Lafayette Intermediate School
Grade: 6
Category: EGSD
Abstract: The purpose of this experiment was to determine which household material design keeps an enclosed space the coolest over time without electricity. This project focuses on passive cooling, which reduces heat using natural materials instead of air conditioning. Four identical cardboard boxes were tested: a plain cardboard box, a reflective box covered in aluminum foil, an insulated box covered in Styrofoam, and an evaporative box covered in a wet paper towel. All boxes were placed in a 37°C heated environment, and temperatures were recorded after five minutes across three trials. The insulated box had the lowest average temperature at 24.3°C. The evaporative box averaged 25°C, the reflective box averaged 28.7°C, and the control box was the warmest at 30.5°C. These results show that insulation was the most effective for slowing heat transfer in this experiment. Although my hypothesis predicted that evaporative cooling would work best, insulation performed better in a closed heat environment. This experiment demonstrates that simple materials can significantly reduce heat gain without electricity and could help improve sustainable cooling methods in the future.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: West Lafayette Jr/Sr High School
Grade: 11
Category: ENBM
Abstract: Cells regulate how they behave in their environment through molecular clutch mechanisms that actively bind and unbind under different types of forces and loads. This property of cells leads to them adjusting their force in response to changes in stiffness and mechanical resistance. Robotic grippers generally lack this ability and instead rely on fixed control strategies that fail to function when object properties differ from their existing settings. This project explores how clutch-based mechanics can build a computational framework for adaptive gripping. A physics-based simulation of clutch dynamics was built in Python based on specific libraries such as NumPy, which leads to modelling each clutch element as a two-state (bound/unbound) system with force-dependent transition rates. Given changes were updated probabilistically, assisted by a Monte Carlo process, while the gripper-object interaction was modelled as springs in series with the utilization of Hookean mechanics. Object stiffness and deformation properties were mapped using metadata in the Yale-CMU-Berkley (YCB) object dataset to preserve mechanical realism. Each simulation produced mechanical response features, which include steady-state force, estimated stiffness, binding fraction behavior, and maximum supported load. To ensure performance differences were not artifacts of stochastic simulation noise, controller comparisons were evaluated using nonparametric inference. Permutation testing was used to assess whether observed endpoint differences could plausibly arise under the null hypothesis of no controller effect; bootstrap resampling was conducted to compute the 95% confidence intervals for mean performance differences. All analyses were performed in Python using fixed seeds to ensure reproducibility. Across 20,000 stochastic trials, the adaptive controller increased mean time-to-slip relative to fixed-force control (, 95% bootstrap CI excluding zero) and showed smaller but consistent improvement relative to PID control. The resultant features were clustered using K-means, which revealed three major, mechanically distinct groups corresponding to soft, medium, and stiff YCB objects. For each given cluster, an optimal grip force was defined as the minimum force required to avoid slip while remaining below the estimated threshold. This adaptive, cluster-based strategy was then compared against two traditional controllers: a fixed-force controller and a PID controller. Performance for all these models was tested across multiple trials using force stability, slip behavior, and statistical comparison of force variance. The ``cluster-based'' adaptive strategy showed fewer varying forces and improved grip stability compared to both baselines. These results indicate that clutch-inspired adaptation can improve grip stability under mechanical uncertainty without requiring labeled object models.
Elevator Pitch Video:
School: Marian High School
Grade: 9
Category: BEHA
Abstract: Boundary extension is something people’s minds do when looking at pictures of objects. Instead of seeing and remembering a picture exactly as it looks, people extend the borders and fill in parts beyond the edges of the image (Intraub & Richardson, 1989). Research has shown that boundary extension is greater when pictured objects are perceived as closer, but it is unclear how other features like the number of objects in a picture contribute to the effect. In this study, a publicly available large dataset (Bainbridge & Baker, 2020) was used to examine how subjective distance and number of objects relate to boundary extension. Results replicated the subjective distance effect: the closer the objects were perceived, the greater the boundary extension. The original study was extended by comparing the original object counts to two new methods of counting objects (human and GPT-5.2). Although the different counting methods were not strongly correlated, they all showed greater boundary extension for images with fewer objects. Further analysis showed that subjective distance and number of objects are confounded in natural images. When one variable was held constant, subjective distance continued to show a strong association with boundary extension, but number of objects did not. These findings suggest that subjective distance plays a stronger role in boundary extension than number of objects. A future study should generate new images where subjective distance and number of objects can be experimentally manipulated independently to test their effects.
Elevator Pitch Video:
School: Marian High School
Grade: 11
Category: ETSD
Abstract: In aerodynamics, surface roughness plays a major role in performance, especially at low speeds. This project shows how different surfaces affect the lift-to-drag ratio of a small airfoil in a low speed, homemade wind tunnel, which operates at approximately 13 mph. Four identical NACA 0012 airfoils were tested, each with a unique surface: dimpled, smooth, rough 3D printed, and dimpled. Each airfoil was tested at angles of attack of 0, 4, 8, and 12 degrees. The Lift and Drag forces were measured using calibrated load cells, and multiple tests were conducted for each condition at each angle of attack. The results were computed and graphed. The results showed just how much surface roughness really affects aerodynamic performance. The dimpled and smooth surfaces produced the highest lift-to-drag ratios, particularly near 8 degrees angle of attack. The sandpaper surface showed consistently the lowest efficiency. Each surface showed expected aerodynamic trends, even showing the decrease of performance at higher angles due to the increasing of drag. The data suggests that surface texture and finish plays a major role in low-speed airfoil performance and supports the importance of smooth or dimpled surfaces in small aircraft and drone designs.
Elevator Pitch Video:
School: Carmel High School
Grade: 10
Category: CBIO
Abstract: Triple-negative breast cancer (TNBC) is a clinically aggressive and heterogeneous form of cancer with highly variable responses to neoadjuvant therapy. TNBC makes predicting pathological complete response (pCR) even more challenging than other cancers, given its negative responses to three key receptors: HER2, estrogen, and progesterone. While pCR is strongly associated with improved prognosis, the relative contributions of tumor-intrinsic biology and the tumor immune microenvironment to treatment response remain blurry. Our study compared predictive models for pCR in TNBC and identified molecular, stromal, and immune features most strongly associated with therapeutic outcomes. We analyzed TNBC patient datasets from cBioPortal with annotated molecular subtypes, tumor characteristics, immune infiltration metrics, stromal composition, and PD-L1 expression. Distinct patterns in pCR probability emerged across molecular subtypes. After experimentation, HER2-enriched tumors showed strong associations with pCR consistent with increased chemosensitivity. Higher stromal scores were negatively associated with pCR, suggesting that stromal-rich tumors may reduce therapeutic effectiveness. Normal-like and Luminal B subtypes demonstrated smaller effects consistent with lower proliferative activity, while immune-associated features trended toward higher pCR probability, supporting the role of immune engagement in response to neoadjuvant therapy. Overall, our findings indicated that pCR in TNBC is shaped by both tumor-intrinsic subtype biology and the surrounding microenvironment. Incorporating stromal and immune-related characteristics improved predictive modeling and provided a more comprehensive understanding of treatment response variability. We experimented on five models using consistent features to predict pCR outcomes, including random forest, logistic, gradient boosted regression, neural network, and support vector models. Our logistic model provided regressions based on both numerical and categorical variables, with one-hot encoding and feature scaling applied to ensure model interpretability, resulting in high contributions from the features PAM50 Subtype HER2, TNBC Subtype Basal-Like 2, TNBC Subtype, Immunomodulatory TNBC Subtype unknown, and stromal score. The random forest model served as an ensemble-based approach as it integrated features based on multiple decision trees to derive feature importance with impurity reduction, providing results featuring mutation count, PDL1 Combined Positive Score, CD3 Positive IHC (%), stromal score, and tumor purity having the highest importance. Using a gradient boosted regressor allowed us to eliminate errors and maintain a clearer picture of pCR interactions, with mutation count, PDL1 Combined Positive Score, CD3 Positive IHC (%), stromal score, and immune score having high importance. Our support vector machine acted as a supervised machine learning algorithm to help as a classification model for complex biological data and resulted in CD3 Positive IHC (%,) stromal score, PDL1 Combined Positive Score, mutation count, and TNBC Subtype Unspecified having high importance. Finally, our deep neural network used its artificial neurons in multiple hidden layers for predictive modeling to derive complex relationships, citing the strongest relationships in the features immune score, CD3 Positive IHC (%), stromal score, PDL1 Combined Positive Score, and mutation count. Our study utilized the receiver operating characteristic area under the curve (ROC AUC) in order to provide more accurate results and analyze the positive correlation between our PAM50 subtypes, PD-L1 expression, stromal composition, tumor purity, and pathological complete response. Our logistic regression model achieved a score of 0.68, the highest score in comparison to the four other models we experimented with, meaning it has greater capacity in predicative power. The different machine learning models used different methods of ranking feature importance, but some features remained significant across the models, such as the mutation count, stromal score, and PAM50 subtypes. We demonstrate that pCR results in TNBC react positively to tumor subtype and microenvironmental context, in tandem with multiple factors influencing some features’ stronger contributions relating to the prediction of pCR. These factors, paired with our analytical models, further TNBC pCR research and earlier diagnosis to aid in prevention and patient care.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Penn High School
Grade: 9
Category: PLNT
Abstract: This study examines how Wifi, a non-ionizing radiation operating at radiofrequencies of 2.4, 5, and 6 GHz, affects the growth and development of living organisms. Specifically, two plants, Raphanus Sativus (Cherry Belle Radish) and Lepidium Sativum (Garden Crest) are exposed to the most common Wifi devices in use today, namely, a Wifi booster, iPhone, and iPad. These devices are ubiquitous–found in almost every public and private space –and are nearly inseparable from people in the modern world. Yet, the effects of Wifi on living organisms are not well known and remain obscure. In this experiment, the plants were physically separated into four groups under the same environmental conditions (sunlight, water, temperature, soil) with three groups exposed to a Wifi booster, iPad, and iPhone, respectively, and one group without a device serving as the control. Observations of growth and measurements of plant height were taken and recorded daily. The results demonstrate that the germination of Raphanus Sativus occurred more rapidly when subjected to Wifi radiation, especially from the Wifi booster. In addition, the seeds of this plant were more likely to germinate in all three groups subjected to Wifi than in the control group. In general, this plant appeared to thrive more when exposed to Wifi, showing a more effective and rapid seed germination, larger overall leaf widths, and taller plants during development. The effects of Wifi on Lepidium Sativum were less pronounced, as all groups appeared to germinate and grow similarly, demonstrating that Wifi affects plants differently.
Elevator Pitch Video:
School: Marian High School
Grade: 11
Category: BMED
Abstract: Effect of Lipopolysaccharide on Antibiotic Susceptibility in A. Baumannii Paloma Rizzi, Marian High School, Mishawaka IN, United States of America How does the absence of LPS affect antibiotic susceptibility in A. baumannii? A. baumannii is one of the few gram-negative species that can survive without lipopolysaccharide (LPS), a permeability barrier that forms in the outer membrane. It was hypothesized that the mutant (AB-84R) strain will be more susceptible to daptomycin and clindamycin than the wild-type (AB-84) strain due to the absence of LPS, and the opposite for colistin because it binds directly to LPS as its mechanism of action. AB-84 and AB-84R strains were cultured and plated onto agar plates. Antibiotic E-test strips were applied to each plate and incubated 18-24 hours at 37°C. MIC values were recorded and averaged across six trials. Statistical analysis was performed using Welch's T-test and fold-change. Results for daptomycin and clindamycin suggested that AB-84R was more susceptible than AB-84. Daptomycin was the most resistant to the AB-84 strains with an MIC average of 256ug/mL, while its AB-84R strain average was 3.67 ug/mL. Clindamycin had an AB-84 average of 219 ug/mL and an AB-84R average of 1.42 ug/mL. Colistin’s results only supported the hypothesis in two trials, while all other trials demonstrated increased susceptibility to AB-84R. Colistin’s AB-84 average was 171 ug/mL, and its AB-84R average was 3.83 ug/mL. This research demonstrates the importance of LPS’s function of regulating antibiotic entry. Understanding LPS’s role is necessary for developing effective antimicrobial treatments.
Elevator Pitch Video:
School: Marian High School
Grade: 11
Category: ENBM
Abstract: Non-invasive detection of disease biomarkers has the potential to transform early cancer screening and long-term health monitoring. Diabetes affects over 400 million people worldwide, while lung cancer and liver cancer are among the leading causes of cancer-related death. Early and accessible monitoring could improve outcomes for these populations. This project presents the development of an Airborne Biomarker Localization Engine (ABLE), designed to collect and analyze biomarkers from open air. The device was designed using SolidEdge 3D and fabricated with a PLA 3D printer. ABLE operates by drawing ambient air into a chamber where it mixes with water vapor and condenses on a cooled plate. The condensed liquid then flows into a collection reservoir for analysis. To validate performance, experiments were conducted using controlled concentrations of glucose and ethanol solutions as model biomarkers. Twelve glucose concentrations were tested and analyzed with commercial glucose test strips. Color changes were quantified using CIELAB color space analysis to objectively measure intensity differences based on concentration. Three ethanol collections were also performed to evaluate the device’s capability to collect volatile organic compounds. Evaporation rates of biomarkers were studied with an optical contact angle goniometer to compare physical properties that influence collection. Results show that the redesigned ABLE device successfully collects measurable samples and distinguishes concentration differences through quantitative analysis. These findings support the feasibility of condensation-based biomarker collection for non-invasive monitoring. Future work will focus on enhancing efficiency, adding real-time sensing, and expanding to clinically relevant biomarkers for broader applications.
Elevator Pitch Video:
School: Fishers High School
Grade: 10
Category: CBIO
Abstract: Antimicrobial resistance is one of the most significant risks that that health industry faces today. Antibiotics are considered crucial, very common forms of treatment for a wide variety of diseases and ailments. However, just as we adapt and evolve to resist the effects of viruses and bacteria, microbes also grow accustomed to the treatments we use to get rid of them. Over time, the treatments are rendered completely ineffective. While we can try and use new treatments that are stronger, scientists are worried that our current methods, which are designed to reduce cost as much as possible, are too dependent on standard antibiotic protocols and will lead to quick efficacy loss. In this project, I use a large database Mendeley to train a machine learning model to identify a series of key points about AMR in clinical scenarios. Using some basic information of the patient and a resistance gene biomarker, my model was trained to identify which of the most common antibiotics for a given disease will be most effective. The data included patient age, gender, the presence of resistance genes, and the susceptibility to a set of antibiotics. Then, the model was tested by giving it the patient data and resistant genes, and it attempted to correctly output the susceptibility to the common antibiotics. The model achieves a high accuracy and can be used in clinical settings to give doctors guidance on which one to use (given other factors are constant). This is important because not only are patients less likely to stay sick or worse because they will be more likely to receive the correct antibiotic, but also, this guarantees that only microbes who have less exposure to such antibiotics are used, slowing down the overall process of resistance development. In the future, more genes, diseases, and antibiotics should be tested to expand the optimal usage of antibiotics and, using the data found, we can design, perhaps utilizing more AI models or Monte Carlo simulations, full cycling routines using combinations of antibiotics that will be the least likely/take the most time to develop any resistance.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: DeKalb High School
Grade: 12
Category: ROBO
Abstract: Chuck, a cornhole-playing robot. Problem statement - How can an industrial robot be mechanically designed and programmed to consistently throw a cornhole bag into a target from a fixed distance? Hypothesis: If a FANUC industrial robot is mounted securely, equipped with properly designed end-of-arm tooling, and programmed with repeatable motion paths, it will consistently throw cornhole bags toward the target with high accuracy. Our project demonstrated that an industrial FANUC robot can be adapted to perform a recreational task such as cornhole with high accuracy. This project highlights how industrial robots can perform non-traditional tasks when hardware and software are carefully designed and tuned. After testing and removing unnecessary variables (the inconsistent bean bags), Chuck had reached 94% accuracy.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: DeKalb High School
Grade: 9
Category: PHYS
Abstract: I made a wind tunnel to test different angles of attack of a wing to see at what angel makes the most air disturbance.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: DeKalb High School
Grade: 9
Category: ENEV
Abstract: As chicken farms increase in numbers in America, so does chicken manure production. With manure comes ammonia, the chemical that not only smells bad but can be harmful to people and chickens. This led to the purpose of my project, which was to economically reduce ammonia levels in chicken manure to result in healthier chickens and lower costs. The methods of this project were testing the starting Parts Per Million (PPM) of the chicken manure for 6 variables and control. Test three trials for each variable and control, each with one pound of chicken manure then, add 1 cup of each variable to the correct container. Then measure the PPM of all 21 trials after 1 hour, 7 hours, 1, 2, 3, 4, 5, and 6 days after starting. My hypothesis is that the 100% zeolite variable will decrease the ammonia the most compared to the control and maintain a level that is healthy for chickens and people. The results showed that the Zeolite variable did the best compared to the control. The microporous arrangement of the Zeolite variable allowed it to absorb the NH_4^+. In conclusion, the zeolite variable was the best ammonia reducer for chicken manure since it was the cheapest, most effective, and has no negative effects on most crops. This will entice farmers to use chicken manure over synthetic fertilizers and encourage chicken farmers to use zeolite as an ammonia reducer for chicken manure. This will result in better health for chickens, fields, and people.
Elevator Pitch Video:
School: Islamic School of Indianapolis, MTI
Grade: 8
Category: BMED
Abstract: This study investigated the effectiveness of various over-the-counter antacids in neutralizing stomach acid. The main objective was to determine which brand out of Milk of Magnesia, Picot, Tums, or Pepto-Bismol, that requires the most hydrochloric acid (HCL) to reach a neutral pH of 3 to 4 which is the normal stomach resting PH. To conduct the experiment, recommended dosages of each antacid were dissolved in 4 separate 200ml of water. Then HCL was added drop-by-drop to each solution of antacid. The pH levels were monitored continuously using a digital pH meter until neutralization was achieved. The results showed that Tums (calcium carbonate) neutralized the highest volume of hydrochloric acid, supporting the initial hypothesis. The null hypothesis was rejected, as significant differences were observed between the brands. While the experiment successfully identified the most potent neutralizer in vitro, limitations included potential pH meter margin of error and the exclusion of biological variables such as stomach enzymes and digestion speed.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: West Lafayette Jr/Sr High School
Grade: 8
Category: ENEV
Abstract: Flooding is one of the most damaging natural disasters in the United States, affecting lives, homes, and communities each year. Accurate flood maps are essential for understanding flood risk and protecting people and properties. However, effective and accurate risk maps are rare. This project uses remote sensing and machine learning to examine flood risks along the Wabash River in Tippecanoe County, Indiana. The objective of the project is to see if satellite radar images from Sentinel-1, along with machine learning and Geographic Information Systems (GIS), could be used to accurately map flooded areas and estimate the number of people and buildings affected. Unlike regular remote sensors, Synthetic Aperture Radar (SAR) uses radar, which can identify water features at night and through clouds, conditions which are common during flood times. Therefore, pre- and post-flood SAR images from Sentinel-1 were compared to detect water spread changes. Several features from the satellite data were used to train a Random Forest machine-learning model to classify flooded and non-flooded areas during a recent flood event in February 2022. The model achieved extremely high accuracy (of over 99%), showing strong reliability. The final flood map was combined with building data and U.S. Census population information using GIS. Results showed multiple flooded buildings and others at risk nearby, with an estimated 1,200 people affected across several census areas. Overall, this study shows that combining SAR imagery, machine learning, and GIS can be highly effective in flood mapping for emergency planning, reducing risk, and saving lives.
Elevator Pitch Video:
School: West Lafayette Jr/Sr High School
Grade: 11
Category: TMED
Abstract: Malaria is among the earliest documented infectious diseases in humans and remains a major global health challenge. In malaria-endemic regions, anemia substantially worsens outcomes, and severe malarial anemia is a medical emergency in children and pregnant women with parasitemia. In holoendemic settings, more than half of deaths occur within the first 24 hours of hospital admission, underscoring that delays in recognition can be fatal. Yet malarial anemia is often missed because rapid, noninvasive point-of-care tools for fast-track detection are limited. In resource-limited settings, separate malaria and anemia tests can fail to flag an imminent emergency and miss the narrow window for life-saving transfusion. Here, we present spatial and textural (radiomic) analyses of smartphone photographs of the palpebral conjunctiva (inner eyelid), coupled with hyperparameter-optimized neural networks, to directly detect malarial anemia noninvasively. This approach leverages readily accessible microvascular changes in peripheral tissue using images captured with unmodified smartphone cameras, offering practical advantages including ease of acquisition and direct visualization of microvasculature. Using data collected in malaria-endemic regions of Rwanda, our radiomics-based model demonstrates reliable performance in distinguishing malarial anemia from non-infected healthy cases. A supervised model optimized for direct malarial anemia detection outperforms an AND-gate combination of independent malaria and anemia detection models. Conjunctival smartphone imaging may enable scalable screening and support rapid triage and management of malarial anemia in resource-limited settings in sub-Saharan Africa.
Elevator Pitch Video:
School: West Lafayette Jr/Sr High School
Grade: 9
Category: EAEV
Abstract: Motor vehicle exhaust, industrial facilities, and wildfires are all affecting the air quality around us. Such processes emit pollutants such as particulate matter (PM10 and PM2.5), ozone (O₃), and nitrogen dioxide (NO₂). Particulate matter's compound, such as PM10, is usually from vehicles, while PM2.5 is influenced by chemical reactions. O₃ is formed through the reaction of gases in the presence of sunlight and is a part of photochemical smog. NO₂ is released from fuels used in industrial and transportation areas. This project examines three different machine learning (ML) models to predict air pollution as a function of local pollution-affected processes. The focus is on predicting the air quality using weather and traffic data. Using weather and traffic data could establish a direct way to predict pollution in real time using information such as how many cars are currently on the road. Weather also affects air quality because rains can wash pollutants out and wind can move pollutants away. My project aims to identify the relationship between these different variables using 3 different ML prediction models, which are compared to each other in their efficacy. The 3 different models include Gradient Boosting, K-Nearest Neighbor, and Random Tree. All three of these models have different ways of predicting air quality levels. They are compared by metrics such as RAE, RMSE, MAE, and R² for error and using a confusion matrix for their precision.
Elevator Pitch Video:
School: West Lafayette Jr/Sr High School
Grade: 8
Category: MCRO
Abstract: 1. Introduction: Yeast serves as a microscopic biological engine that converts carbohydrates into energy through fermentation. This process is a cornerstone of sustainable biofuel production, offering a renewable alternative to fossil fuels. 2. Problem Statement: This study investigates how the structural complexity of carbohydrates, monosaccharides, disaccharides, and polysaccharides, affects the rate and efficiency of yeast metabolism and CO2 production. 3. Procedures: Using a syringe-based system at a constant 35°C, CO2 output from glucose, sucrose, lactose, and starch over ten minutes was measured. An additional group of starch treated with α-amylase was tested to observe the impact of breaking down complex molecular chains. 4. Results: Glucose (monosaccharide) exhibited the highest efficiency, exceeding the 10 mL syringe capacity within 8 minutes. Sucrose (disaccharide) followed with a steady rate, while starch alone showed negligible reaction. However, starch treated with α-amylase significantly increased gas production. Despite its disaccharide structure, lactose produced almost no CO2 due to its specific molecular linkage. 5. Conclusions: Yeast produces gas faster with simple sugars because their basic structure is easier to process for energy. Complex carbohydrates must be broken down by enzymes into smaller pieces before fermentation can happen. These findings help us choose the best raw materials to make future biofuel production more efficient.
Elevator Pitch Video:
School: Carmel High School
Grade: 12
Category: CBIO
Abstract: Lung cancer remains the leading cause of cancer-related mortality worldwide, and early detection significantly improves survival outcomes. Subsolid pulmonary nodules, including pure ground-glass and part-solid nodules, often represent early-stage adenocarcinoma and exhibit distinct radiographic and biological characteristics compared to solid nodules. Although deep learning-based AI systems have demonstrated high overall sensitivity in automated lung nodule detection, performance is typically reported as a single aggregate metric without stratification by nodule morphology. This study evaluated whether a 3D Convolutional Neural Network (CNN) demonstrates morphology-dependent disparities in detection sensitivity using the LIDC-IDRI dataset. Over 1,000 CT-derived 3D nodule volumes with four expert annotations per lesion were analyzed. Nodules smaller than 3.0 mm were excluded and images were normalized to preserve contrast differences between solid and ground-glass nodules. Data were split 80% for training and 20% for testing with patient-level separation to prevent data leakage. On independent testing, the model achieved 80% overall accuracy and 93% sensitivity (25 true positives, 2 false negatives, 15 true negatives, 8 false positives). However, stratified analysis revealed the highest detection sensitivity for solid nodules, intermediate performance for part-solid nodules, and the lowest sensitivity for ground-glass nodules. Multivariable logistic regression demonstrated that morphology remained a statistically significant independent predictor of detection failure (p < 0.05). These findings indicate that despite strong overall performance, the model exhibits morphology-dependent detection disparities, with potential implications for equitable early-stage lung cancer identification.
Elevator Pitch Video:
School: Terre Haute South Vigo High School
Grade: 11
Category: CBIO
Abstract: Prolonged spaceflight presents unique physiological challenges to the human body, many of which stem from the induced cephalad fluid shift that lead to reduced renal blood flow. This project investigates the molecular mechanisms behind sodium retention and fluid overload in microgravity, processes that remain poorly characterized at the molecular level, as specific pathways in this response are not well understood. Preliminary data from the NASA Open Science Data Repository dataset OSD-513, a transcriptomics study of Mus musculus kidney tissue in long-term spaceflight, revealed significant overexpression of genes in the 20-Hydroxyeicosatetraenoic Acid (20-HETE) metabolic pathway, specifically genes that coded for Cyp4a14 and Ugt1a enzymes, as part of the Cytochrome P450 (CYP) arachidonic acid metabolism pathway. Its downregulation in microgravity suggests that enhanced breakdown and elimination of 20-HETE may drive sodium retention due to the important role of this molecule. Building on this, the role of 20-HETE, a known inhibitor of sodium reabsorption, led to the central hypothesis that weightlessness promotes active metabolism of 20-HETE to favor sodium conservation. To test this, two follow-up experiments are proposed. First is the characterization of the temporal relationship between the expression of Cyp4a14 and Ugt1a and 20-HETE metabolite levels in mice during spaceflight. A simulated microgravity model would be used to determine whether pharmacological inhibition of 20-HETE metabolism can mitigate sodium retention. The findings of this study will provide critical insight into the kidney’s adaptive response to microgravity and may identify novel therapeutic targets for reducing the health risks of spaceflight-induced fluid shifts, ultimately improving astronaut health during long-duration missions.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Terre Haute South Vigo High School
Grade: 9
Category: PHYS
Abstract: As we continue to launch satellites into orbit around the Earth, the risk of close conjunctions and even collisions between satellites and debris steadily grows higher. This may lead to a dangerous scenario in which multiple collisions could cause a cascading effect, damaging or destroying many or most satellites in orbit and making Lower Earth Orbit virtually unusable. This currently hypothetical scenario was termed "Kessler Syndrome" in 1978 by Donald Kessler and Burton Cour-Palais. In this project, I attempted to learn more about the effect high-profile constellations, such as SpaceX's Starlink or Amazon Leo (formerly known as Project Kuiper) can have, especially as constellations swell to thousands of satellites. So, using data from space-track.org, a website run by USSPACECOM, I wrote a program using the coding language Python, with some debugging assistance from ChatGPT, to examine every satellite that space-track.org provides data on and analyze its orbit. Each satellite's orbit was compared with approximately 30,000 tracked objects in orbit and conjunctions, meaning close encounters between satellites, were analyzed. The variability of each satellite's orbit was closely approximated using estimated covariance data based on historical satellite information. From there, using the method proposed by Salvatore Alfano in 2005, I calculated the probability that each pair of conjoining satellites would collide.
Elevator Pitch Video:
School: Penn High School
Grade: 9
Category: BMED
Abstract: Over 389,000 people develop colon/lung cancer annually. This project investigated how Cisplatin chemotherapy treatment on colon/lung cancer cells impacted gene expression for GATA4, FOXD2, ZNF558, RNF8, and NR5A1. Results will determine if any selected genes are potential biomarkers for Cisplatin treatment; identifying biomarkers can improve cancer treatment/outcomes. It was hypothesized that in colon cancer GATA4/ ZNF558 will decrease, FOXD2/ RNF8 will increase, and NR5A1 will show no change in expression and in lung cancer, the expression of the GATA4/FOXD2/ ZNF558 will decrease, RNF8 will increase, and NR5A1 will show no change in expression. Colon/lung cancer cell lines previously treated with Cisplatin had their cDNA frozen. Thawed cDNA from these cells was used and the expression of GATA4, FOXD2, ZNF558, RNF8, and NR5A was measured. The data was amplified by a qPCR machine; the mean was calculated using the delta-delta CT method. T-Tests to determine statistical significance were performed. Utilized prior data of gene expression in blood cell lines to interpret trends of sensitivity or resistance. In colon cancer cell lines, FOXD2 is a potential Cisplatin resistance biomarker while ZNF558 and RNF8 are potential sensitivity biomarkers. All RNF8 timepoints are statistically significant (p <0.05). GATA4 is not expressed in the colon. In lung cancer cell lines, 2 timepoints for 3 genes were potential Cisplatin biomarkers. FOXD2 and GATA4 align with potential resistance patterns and ZNF558 align with potential patterns of Cisplatin sensitivity. Select gene timepoints were statistically significant (FOXD2 18h p <.01; ZNF558 6h p <.01 and 18h p <.1).
Elevator Pitch Video:
School: Carmel High School
Grade: 9
Category: BMED
Abstract: The primary objective of my project is to identify tumor-specific antigens using computational simulations for the design of a peptide-based cancer vaccine targeting neuroblastoma. I use the simulations to predict the 3D models of parts of the Neuroblastoma (epitopes) to see which binds to the B-Cell receptor the best. This shows the specific antigens for the peptide vaccine.
Elevator Pitch Video:
School: Lafayette Jefferson High School
Grade: 11
Category: CHEM
Abstract: The process of making transportation fuels currently is consumes a high amount of energy as well as it leaves behind toxic byproducts. The current way of making fuel is by refining crude oil. However, there is a way to make gasoline from natural gas by using C3 and C2 hydrocarbons. This process uses catalysts to make the large hydrocarbon chains that fuel consists of. These chains contain anywhere from 8 to 12 carbon atoms. In this experiment, catalysts are tested to figure out what kind works best and what type of active site is working. This will be tested by constructing a catalyst, then placing the catalyst under heat to make hydrocarbon chains via oligomerization and olefin cracking. Infrared Spectroscopy will also be used to properly pinpoint what active sites are on the catalyst compared to that of zeolites. It has been shown that immobilized Td Ga3+ and Fe3+ ions on amorphous silica have shown to be Lewis acidic contrary to those seen in Zeolites. It is hypothesized that immobilized Td aluminum ions on amorphous silica will form Lewis acid sites similar to ones found in zeolites and be active for hydrocarbon reactions. The results of this experiment showed that the catalysts made had Lewis active sites and they preformed the reactions that yielded the same products as that of zeolites. The chemical reactions done are different for the zeolites and our aluminum silicate but they resulted in the same hydrocarbon chains to be made. These results will help improve knowledge on making fuel from small hydrocarbons chains into large fuel hydrocarbon chains. If this process of making larger hydrocarbons can become a reliable low energy way of making fuel it will cause there to be no toxic byproducts and it will use less energy than is currently used to make fuel.
Elevator Pitch Video:
School: Carmel High School
Grade: 12
Category: ENEV
Abstract: As the most used man-made material, concrete is responsible for 8% of global CO2 emissions. The process of carbonation immobilizes CO2 into stable carbonates via dissolution and recombination with magnesium or calcium. This presents a dual opportunity when certain concrete aggregates are carbonated during curing: storing emitted CO2 in the concrete matrix while enhancing material properties. This research utilizes accessible aggregates to develop a structurally sound concrete mix that effectively sequesters CO2 upon carbonation curing. The concrete mixes were designed using an L9 Taguchi Array with 4 factors and 3 levels, creating 8 total mixes. Sequestration ability was measured by comparing samples cured in a pressurized chamber to identical samples cured in the ambient environment to determine the mass of CO2 gained. Structural strength was determined through compression testing, comparing experimental samples versus control samples and industry standards. Each experimental sample captured between 2 and 5 grams of CO2 per 28.3 cm^3, with the best mixes capturing 5.9% of their total mass. All experimental samples withstood higher compression than their control counterparts, with 3 mixes surpassing industry standards for commercial use. Based on the initial data, regression was used to design an optimized mix, the samples of which captured 6.4 grams of CO2 (6.6% of their original weight). By developing mixes that both sequester atmospheric CO2 into concrete and gain compressive strength, this project advances a path towards a scalable and sustainable method of concrete production and a circular carbon economy.
Elevator Pitch Video:
School: Carmel High School
Grade: 11
Category: CHEM
Abstract: Aryl Halides are significant compounds that have great potential in cross-coupling reactions and interconversions of functional groups. At a larger scale, they play a role in the development of agrochemicals, pharmaceuticals, and materials manufacturing. To functionalize these groups, the most current approaches are with metal catalysts. However, a direct activation of C(sp2)-X through photochemistry is lacking because of the limiting negative reduction potential that Aryl halides possess. Therefore, it becomes imperative to evaluate NHC’s validity as a photocatalyst for boronation functionalization because of their ability to become powerful electron donors when excited under light. Thus, a mild, photo-redox-catalyzed protocol for the oxidative coupling of aryl halides with boronic esters is reported, enabled by N-heterocyclic carbenes (NHCs) acting as stand-alone photocatalysts. This transition-metal-free approach eliminates the need for exogenous photocatalysts, ligands, or thermal activation, offering a sustainable and operationally simple alternative to conventional methods. Our system underscores the untapped photochemical potential of NHCs in driving single-electron transfer (SET) events under visible light irradiation. By leveraging their inherent electron donating ability and persistent carbene character, this methodology expands the reactivity space of NHCs beyond traditional umpolung catalysis involving Breslow intermediates. This work contributes to the exploration of carbene-centered photoredox chemistry and establishes NHCs as viable photosensitizers for photoredox-catalysis.
Elevator Pitch Video:
School: Lafayette Jefferson High School
Grade: 11
Category: BCHM
Abstract: The formation of insoluble amyloid fibrils from proteins is responsible for many different diseases, including Alzheimer’s and Parkinson’s disease. The ability to label these fibrils has the potential to be able to detect these diseases. In my experiments, I developed probes (Z compounds) to label protein amyloid fibrils, such as those from hen egg white lysozyme (HEWL) and alpha-synuclein. Human lysozyme can form amyloid deposits in the liver leading to a fatal disease, and alpha-synuclein aggregation can lead to the formation of Lewy Bodies leading to Parkinson’s disease and dementia. Protein aggregation to form fibrils involves burying hydrophobic groups of the protein inside of the fibrils to form a hydrophobic core. I hypothesize that the more hydrophobic the components are within the probes I develop, the better they will detect the amyloid fibrils by fluorescence. I found that my hypothesis was correct proving that the more hydrophobic the compounds were the greater the increase in fluorescence with the HEWL amyloid fibrils. I also found that I could use these probes to rapidly monitor the dissolution of the HEWL amyloid fibrils. Finally, I found that two of the Z compounds were also able to fluoresce with alpha-synuclein fibrils, showing the utility of these compounds to probe two different protein fibrils. Overall, two compounds (Z-61 and Z-62) worked better than the known thioflavin T (ThT) with the HEWL fibrils, and one compound (Z-100) performed as well as ThT with two different amyloid fibril samples. Ultimately this work could provide new probes to monitor diseases based on amyloid fibrils formation from proteins.
Elevator Pitch Video:
School: Lafayette Jefferson High School
Grade: 12
Category: ETSD
Abstract: Formula 1 is the most prestigious motorsport competition, with 10 teams competing on a track circuit using high-performance cars. Points are given to drivers that place high enough, with Grand Prix races lasting between one-and-a-half hours to two hours (Formula 1,2025). Drivers use their skill and strategic thinking to overtake. The points of interests, such as corners, on an F1 circuit are strategic areas for overtaking as the downforce created by the high G-forces experienced in high-speed cornering are exploitable and uncertain. As “downforce has a direct impact on road adhesion and cornering speeds,” a driver must properly assess how the grip will be affected by track conditions the nature of the POI (Downforce, 2025). Each track is composed of three distinct sectors that encompass different POIs. In instances where drivers misjudge an overtaking attempt or external factors unpredictably affect the cars grip, collisions occur. If trends of crashes and overtakes on F1 race circuits can be identified in relation to PIOs and sectors, then necessary areas of improvement can be found.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Culver Academies
Grade: 11
Abstract: The application of Unmanned Aerial Vehicles in Search and Rescue operations is highly limited due to the low adaptability and durability of conventional UAVs with rigid structural designs. A transformable drone, or Morphobot, therefore exists to resolve this issue through transforming based on the surrounding to better adapt to the environment. Previous solutions such as Multi-Modal Mobility Morphobot, or M4, have provided an approachable path in Morphobot design. (Sihite, E. et al, 2023) However, M4’s joint design connected servos right on the joint axle, causing potential vulnerabilities to external forces when experiencing accidents. This study, therefore, focuses on redesigning the UAV’s joint structure through implementing self-locking multi-link system. Such design improves structural durability as external forces are now redirected to rigid structures rather than servo. Testing results have shown that over 70% of the force has been redirected to structural components that would otherwise damage the servo. The electronics are not experiencing force and are excluded from the force-resisting structure. This redesigned joint structure shows significant improvement in structural durability and is helpful to make M4 more durable in SAR operations. The improved durability of M4 allows the UAV to search into collapsed buildings without concerns on structural failure due to falling rocks or other accidents during the operation and ensures the safety of human rescuers as the UAV can be operated remotely to maneuver or transform by commands from the rescuers in safe zones.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Culver Academies
Grade: 11
Abstract: Ankle joint injuries are common and have risks of long-term functional impairment if rehabilitation is not timely and efficient. Traditional rehabilitation methods rely heavily on therapist experience, lack quantitative assessment, and are difficult to sustain in domestic settings. To overcome these limitations, an intelligent assistive diagnosis and rehabilitation device for ankle injuries is developed to improve rehabilitation accuracy, safety, and accessibility. The system adopts a serial multi-degree-of-freedom structure driven by three servo motors whose rotational axes intersect at the anatomical center of the ankle, enabling all six fundamental ankle movements. Torque sensors provide real-time force feedback for motion assessment and safe patient usage, while an embedded controller and wireless communication support data acquisition and user interaction through PC end and mobile phone end self-designed applications. Experimental results show that the device's motion range fully covers normal ankle movement and that calibrated torque thresholds allow responsive and safe assistance. The system demonstrates the ability of combining quantitative evaluation and multimodal rehabilitation in a low-cost solution, helping to engage accessibility from home.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Lafayette Jefferson High School
Grade: 9
Category: ANIM
Abstract: what did you you do, why did you do it, what did you find out and why does it matter?
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Marian High School
Grade: 10
Category: MATS
Abstract: If glass fiber-reinforced thermoplastic polyurethane (GFRTPU) matches or exceeds the creep resistance and thermal stability of fiberglass epoxy resin, then GFRTPU should be considered an alternative material for wind turbine blades. Wind turbine blades are typically manufactured with fiberglass epoxy resin, which presents significant recycling challenges. This research investigated Elastollan® R3000, a recyclable GFRTPU, as a potential alternative material. GFRTPU samples were tested under three thermal conditions (68°F, 120-140°F, 160-200°F) with a ~5.5 oz load to measure creep behavior, and three samples were tested on an electronic tensile testing machine at ~65% ultimate tensile strength (UTS). Comparative analysis with wind turbine blade specifications revealed that Elastollan® R3000 maintains predictable dimensional stability with minimal creep throughout testing. Unlike traditional fiberglass epoxy composites that require complex modeling to predict progressive deformation over 20–25-year service lives, GFRTPU ensures the 23× safety factor remains constant throughout the blade's operational lifetime while enabling full material recyclability at end-of-life. The research supports the hypothesis that GFRTPU has the potential to replace fiberglass epoxy resin in wind turbine blade applications.
Elevator Pitch Video:
School: Trailside Elem School
Grade: 4
Category: EAEV
Abstract: In the summertime, blacktop is very hot. My feet would burn when walking from our pool across the asphalt driveway to the bathroom. I wanted to see what color would be cooler to walk on, and then use that color to paint a sidewalk across the asphalt. The goal of this project was to determine what paint color cools the driveway the best. I hypothesized white will be the coolest color because it reflects most of the sun’s rays. This was a 28-day, two part study measuring surface temperature various paint colors. In part 1, two sections of asphalt were painted with 12 different colors. Surface temperatures of each color were measured using an infrared thermometer three times a day for 14 days. For part 2, a white sidewalk was painted across the asphalt, as white was the coolest color from part 1. The temperature of this newly painted sidewalk was measured daily against the asphalt for 14 days. A total of 1036 temperature measurements were taken for my 28-day study. The average surface temperatures were: black asphalt 124.9F, sealant 124.4F, white 111.1F, topiary tint 114.5F, relish 119.0F, reseda green 120.1F, shamrock 123.0F, swimming 113.9F, ebbtide 117.4F, georgian bay 121.4F, grey screen 114.9F, and earl grey 120.7F. White had the greatest temperature reduction of 13.8F. For Part 2, the average temperature of the newly painted sidewalk was 93.3F compared to blacktop at 114.9F, for a temperature difference of 21.6F. White was the most effective color for reducing asphalt surface temperature.
Elevator Pitch Video:
School: West Lafayette Jr/Sr High School
Grade: 11
Category: SOFT
Abstract: Molecular geometry optimization is a key part of a long drug discovery process that takes over 12 years and has an approximately 90% failure rate. Faster molecular optimization (MO) can significantly speed up drug candidate screening. Computational models, including DFT, perform MO by finding the lowest-energy structure. However, they can have accuracy limitations, such as difficulty correctly describing weak interactions (hydrogen-bonding/van-der-Waals forces). Variational Quantum Eigensolver (VQE) solves this, providing an exponential advantage using quantum computing. Screening with VQE can reduce the number of physical molecules synthesized by 30-50%, potentially shaving 6 to 12 months off the pre-clinical development timeline. VQE implemented on quantum computers can potentially perform complex molecular geometry optimizations that would take thousands of years into ones that take hours or days. However, estimating the Hamiltonian energy required for VQE is inefficient, representing a fundamental bottleneck due to costly and slow measurements. In this project, a Hamiltonian energy estimator, SymQNet, was developed. SymQNet relies on reinforcement learning to strategically maximize the amount of useful information from a single Hamiltonian measurement, leading to lower error compared to baseline estimation methods (HAL-FI and GreedyBED). SymQNet offered flat runtime scaling up to 20 qubits, indicating significant potential to be applied to larger optimizations. In this work, MO on VQE with SymQNet’s energy estimation offered a reduction in VQE shot attempts by 46.7% and a speedup of 1.88x compared to VQE with standard Pauli-Term estimation. Overall, results suggest that SymQNet offers a significant advancement in VQE for drug discovery.
Elevator Pitch Video:
School: Hamilton Southeastern HS
Grade: 11
Category: SOFT
Abstract: Alzheimer’s disease and age-related cognitive decline limit memory engagement and caregiver insight for millions of older adults. This project developed and evaluated RecallLive, an AI-assisted mobile application designed to support structured memory interaction for senior adults. The system integrates three coordinated subsystems within the RecallLive Cognitive Support System (RCSS): (1) the ReLive Visual Composer (RVC), which extracts metadata and applies temporal and spatial clustering logic to generate structured memory videos; (2) the ReLive Response Analyzer (RRA), which employs a convolutional neural network deployed on-device to classify frame-level emotional responses and construct video-specific emotional timelines; and (3) an LLM-based Engagement Report Generator that converts aggregated emotional classification outputs into interpretable summaries for caregivers through a secure dashboard architecture. Evaluation was conducted in two phases. Phase 1 involved a video-based usability assessment with 102 U.S. adults aged 65 and older. Participants completed a structured survey measuring usability, engagement, design, usefulness, and intention to use. All scales demonstrated strong reliability (Cronbach’s α = .88-.94). Engagement (M = 4.16/5.00) and perceived usefulness (M = 4.11/5.00) were rated highly. Perceived usefulness strongly predicted intention to use or recommend the application (R² = .665, F(1,100) = 198.5, p < .001). Phase 2 included hands-on usability assessment with five participants, confirming clarity of navigation and interpretability of emotional feedback. Results suggest that combining AI-driven emotional analysis with structured visual memory exposure is both technically feasible and scalable for real-world use. This study demonstrates how integrating machine learning with user-centered mobile design may support proactive cognitive engagement strategies for aging populations.
Elevator Pitch Video:
School: East Noble High School
Grade: 12
Category: BEHA
Abstract: The experiment was conducted to see if AI can be a useful tool to help with retention and learning. It was hypothesized that neither method of study held an advantage over the other, and there would be no significant difference between the group. The experimental design consisted of two groups. One control group, with no AI, and the other experimental, with AI. Both groups were given a random topic to study. Each group, AI and No AI, was divided into three groups based on a topic. For this test, the three topics were, the Properties of Bismuth, the Romanian Revolution, and the History of Volleyball. Forty students were randomly assigned to a group and topic. They were given 10 minutes to prepare for a quiz on each topic. Some of the subjects used AI and others did not use AI. Test scores were collected and then analyzed via Welch’s independent t-test. Although there was a percentage difference between the AI and Non-AI test groups, it was not at a significant level with p> .05. In conclusion, the data shows no advantage or disadvantage in recalling facts statistically but does show a greater percentage of correct answers using AI.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Fishers Junior High School
Grade: 7
Category: PLNT
Abstract: How do people that live in small suburban areas with almost no fertile soil to grow plants? Turns out most places with limited access to fertile soil don’t have access to homemade plants, so research was conducted to answer that specific question. Before that research, this essay will discuss how this conclusion was reached. In social studies class, the topic was Egypt and its fertile soil. The class also learned about its surplus of water from sources like the Nile. With that in consideration the leading question was created; If the Coriandrum sativum is grown in water instead of soil then the cilantro's growth rate will increase because the roots of Coriandrum sativum will have easier access to water, oxygen, and nutrients therefore enhancing the Coriandrum sativums growth rate.” At first none of the plants grew, not the plants with soil, nor the plants without soil, which caused the question, “Was there too much water or too little?” They were too cold and too close to the window, which caused the temperature to be off, and temperature is a major benefactor in plant growth - which was the reason for no data and the plants’ germination process to be slow. Overall, the hypothesis was refuted. Coriandum sativum with the most soil grew the largest, it might have been because the cilantro with no soil had an influx of water, however this could still be used in climates that have a decrease on good fertile farming soil to grow organic crops instead of store-bought processed plants.
Elevator Pitch Video:
School: Hamilton Southeastern Jr High School
Grade: 7
Category: BMED
Abstract: Does Posture Affect Reading? This experiment was preformed to find put the best posture for reading the most pages in 30 minutes. The problem this experiment was facing was: “Does posture affect the pages read in 30 minutes?” The experiment was conducted by having the participants first read in a sitting position, then recording the completed pages read, then having 2 read in supine position and the other 3 read in standing position, and then recording the completed pages read for that round; the participants would then switch positions to read in their final position, and their pages read would be recorded. Where the Red Fren Grows was the book used in the experiment. The procedure consisted of gathering the participants, having them read in the different positions and recording their pages completed, then doing that again for a second trial, then finding the averages of the pages completed per posture. The hypothesis for the experiment was: “If the absolute value of a degree of a posture increases 90° from 90°, then the pages read will decrease because the unfamiliar posture will distract one from reading.” This hypothesis was refuted because the posture expected to have the highest average, sitting posture, actually had the lowest average pages read; the posture with the highest average page read was actually the standing posture. The average pages read for sitting posture was 28.3; the average pages read for standing posture was 32.4; the average pages read for supine posture was 32.3. The research showed that fatigue caused by unfamiliar positions can cause fatigue, which can cause a distraction from reading. This was the base of the hypothesis, since it was thought that the standing and supine posture would cause more fatigue and distraction than the sitting posture.
Elevator Pitch Video:
School: West Lafayette Jr/Sr High School
Grade: 11
Category: CBIO
Abstract: One in 8 men get prostate cancer in their lifetime. Unfortunately though, single drug treatments for prostate cancer often cause the cancer to develop resistance to these drugs and become a more aggressive form of the cancer. A possible solution to combat this drug resistance is to prescribe patients a mixture of two or more drugs, called combination therapy. However, there are millions of possible combinations for treatment, as the drug space is vast. This means we need a screening method to reduce the number of plausible combinations for scientists to test in labs. This is exactly what my project PROC-DOT does. PROC-DOT uses genetic data to rank drug combinations. It hinges on the idea that better drug combinations should be able to reverse the negative effects of a disease genetically. Using this technique and methods to reduce the dimensions of the dataset, PROC-DOT was able to rank 80,000 drug combinations for prostate cancer in terms of how effective they would be. Numerous combinations that ranked well on PROC-DOT were confirmed by literature. Additionally, PROC-DOT was validated on a lung cancer dataset where it predicted resultant gene expression vectors for combinations with a high level of accuracy. Overall, PROC-DOT serves as an initial screening method for drug combinations, and has the potential to significantly decrease the costs and time spent on finding ideal drug combinations in future lab trials.
Elevator Pitch Video:
School: Islamic School of Indianapolis, MTI
Grade: 9
Category: MCRO
Abstract: This experiment investigates the impact of sound frequencies on the growth of yeast, specifically Saccharomyces cerevisiae. My experiment exposes yeast cultures to an 800 Hz sound wave at 75-80 decibels for 28 hours, comparing their growth to a control group kept in silence. Despite prior research suggesting that sound waves can influence microorganism development, the results indicated minimal to no observable difference in growth between the sound-exposed and control groups. These findings suggest that, within the short incubation period, sound frequency and decibel level do not significantly affect yeast growth. This research contributes to understanding the potential effects of sound on microorganisms and highlights the need for further studies with extended exposure times and varied sound parameters.
Elevator Pitch Video:
School: Carmel High School
Grade: 12
Category: CBIO
Abstract: Drug-induced liver injury (DILI) is one of the leading causes for acute liver injury in the United States, accounting for 13% of all cases of acute toxicity. DILI represents a significant and consistent barrier to new pharmaceutical development and regulation as clinical trial rejections waste resources and cause delays in critical medicine development. Although significant research has been conducted on the viability of machine learning models in predicting DILI through chemical structure analysis, there remains a significant gap in research dedicated to using bimodal approaches involving biological indicators. In this study, we propose HEPAR (Hepatotoxic Ensemble Predictive Analytics Resource), a novel approach to machine learning-based DILI prediction. HEPAR combines traditional chemical structure analysis with bioindicator analysis to create a holistic understanding of DILI. We compiled SMILES (Simplified Molecular Input Line Entry System) strings for 1,199 pharmaceutical compounds and analyzed their topological and chemical fingerprints using the MACCS bit system, while sourcing enzyme-compound interaction data from bioinformatics APIs. Our model uses a diverse cohort of base learners including deep neural architectures, gradient-boosted decision trees, ensemble trees, and geometric classifiers. Our research found that an ensemble-based voting system composed of all base-learners improved performance significantly compared to the individual models across several metrics. HEPAR demonstrated accuracy, MCC, and F1 values that consistently aligned with, and in several instances exceeded, established benchmarks for hepatotoxicity prediction. HEPAR successfully demonstrates the effectiveness of a bimodal approach in DILI analysis and shows promise as the next significant advancement in DILI prediction. Preemptive toxicity prediction and proactive changes in the development process demonstrate potential for reducing the production times of critical medications while greatly mitigating the effects of liver toxicity on public health.
Elevator Pitch Video:
School: Otter Creek Middle School
Grade: 7
Abstract: This student has not yet submitted an abstract.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Terre Haute North Vigo High School
Grade: 12
Category: ETSD
Abstract: 20% of total energy consumption is used to merely overcome friction (Holmberg and Erdemir, 2017). Friction comes from all places, with new rolling bearings accounting for over 420 terawatt hours of consumption (Bakolas et al). Bearings require proper lubrication to achieve peak performance. Ferrofluid, a colloidal suspension of ferromagnetic particles, has been explored as a lubricant due to its ability to be manipulated by magnetic fields but there is very little research in the application of ferrofluid in radial ball bearings. I designed and tested 3 different designs, two of which I replaced some of the bearings with magnets. I planned on observing how the introduction of magnets in the bearings would impact friction under load and at higher speeds. I was able to estimate the lubrication regime of my bearings by using the Stribeck curve and specific film thickness.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Eastern Hancock Elem School
Grade: 5
Category: BEHA
Abstract: Real or Fake Identify the AI The purpose of this study is to see if people can identify AI images. I think today's pre-teens and teenagers will know which image is AI because they grew up with social media and more AI than if you were born before social media and AI was popular. I created an AI quiz using two online sources to obtain 10 AI images, and 10 real images. I surveyed 82 volunteers via e-mail, through school and home. Volunteers provided age and gender anonymously. Volunteers were asked to identify which image was AI. My control was the 10 images I used to quiz volunteer's, and the quiz I created in Google Forms. My independent variable is the volunteer’s age, and the dependent variable is the different age groups, and their ability to identify the AI images. I collected data by using Google Forms, to determine what age group was able to identify the 10 AI images. I used 80 out of 82 participants, because two volunteers did not specify age. I used google sheets, to make a spreadsheet showing volunteers scores and their ages. I was able to make column graphs representing the data I entered. I broke the graphs down to age ranges and their score, showing a teen group, adult group, and a senior group. I used each age group's score to determine the average score for each age group. I collected my data, using a spreadsheet to find all the averages for each age group. The teen group consisted of 32 volunteers, ages 15-18, with an average score of 6.5 out of 10. The adult group consisted of 32 volunteers, ages 20-49, with an average score of 6.7 out of 10. The senior group had sixteen volunteers, ages 50 plus, with an average score of 6.3 out of 10. The average scores did not vary by much, and actually showed the adult group averaged better than the teen group. These findings do not support my hypothesis, and show the adult group averaged the highest out of all three groups. I believe today’s youth should be taught how to identify AI to maintain credibility of information, as AI and technology continues to evolve.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Honey Creek Middle School
Grade: 8
Category: PHYS
Abstract: In this project, I would be measuring the speed of light in different materials using Snell's Law and the index of refraction. My goal is to observe how light reacts as it passes through various media such as glass, water, and liquid turmeric solution.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Honey Creek Middle School
Grade: 8
Category: BEHA
Abstract: Access to emergency services like hospitals and fire stations is essential for public safety. How close you live to these services can determine whether you survive a medical emergency. This study examined whether lower income and higher income neighborhoods in Greater Indianapolis have different levels of access to emergency services. Geographic data was collected for 32 hospitals, 27 urgent care centers, and 72 fire stations across Marion and Hamilton Counties. To find exact locations, I used Google Maps and city databases. Gathering income information for 28 ZIP codes from the U.S. Census Bureau, I measured the distance from each neighborhood center to the nearest emergency facilities. I compared these distances across three income groups: low-income (under $45,000), middle-income ($45,000-$70,000), and high-income (over $70,000). The analysis revealed crucial differences in emergency service access. Low-income neighborhoods averaged 4.2 miles to the nearest hospital, while high-income areas averaged only 2.1 miles, half the distance of low-income. Fire stations showed the same trend. Low-income areas averaged 3.1 miles to fire stations while high-income areas averaged just 1.4 miles. Five ZIP codes have critically limited access, which affect approximately 85,000 residents. This research demonstrates that where you live and your income level significantly affect your access to emergency services in Greater Indianapolis. Lower-income communities must travel farther to reach hospitals and fire stations. These findings support the need for building new facilities in underserved areas and highlight the importance of fair emergency service planning.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Marian High School
Grade: 11
Category: BCHM
Abstract: A bioassay is required to assess the potency of a novel drug. A whole-animal potency test can be developed using a generated strain (tgMORr) of the nematode C elegans. The strain contains the human mu-receptor, allowing them to respond to opioids. Before working with them, brine shrimps will be used as a model to develop methods for handling small organisms. To do so, it is important to observe their behavior in a controlled environment and how they react to being dosed with drugs. My hypothesis is that if brine shrimp are dosed with drugs, then their speed and behavior will change due to having receptors that can detect changes in their environment. Firstly, brine shrimps were raised to adults and dosed with drugs (caffeine, alcohol, procaine, xylazine), with all–except alcohol (0.1, 0.5, 1.0%)–having the treatments (.25, .5, 1.0mg/ml), each being videoed for 30sec at 10min intervals for 40min. The videos were analyzed via ImageJ for movement and speed. Compared to the control, caffeine-treated brine shrimp showed similar or higher speeds. Alcohol-treated and procaine-treated displayed similar or decreased speeds. Xylazine-treated showed decreased speed, having the most variability out of them all. Overall, these results support that the drugs do induce a behavioral effect on brine shrimps. By doing this project, it provides a foundation for bioassays using C. elegans by developing methods to handle small organisms. In the long term, this will provide a relevant method on testing the potency of novel drugs.
Elevator Pitch Video:
School: Marian High School
Grade: 11
Category: PHYS
Abstract: Photometric measurements of the eclipsing binary star system V808 Aurigae show that its orbital period is changing. Past research has suggested that these changes are due to an exoplanet. Using photometry, I tested my hypothesis that V808 Aurigae has a planet orbiting it. Analysis was performed using photometric data collected by Professor Peter Garnavich of Notre Dame. Python software was used to calculate mideclipse times and orbital periods and perform statistical analysis of the system. Subsequently, data was compared to an ephemeris found in Schwope et al. (2015) and other data from Fig. 4 of Leichty et al. (2024). The average orbital period of the system was 0.08138135239274998 JD, approximately 2 hours (SD = 2.573496933694353 x 10-6 JD, n=2). The mean O - C value of the eclipse (calculated from the ephemeris in Schwope, et al.) was approximately 59 seconds (SD = 1.7192730143841273 sec, n=6). Results are compatible with those in Fig. 4 of Leichty et al., where, after approximately 2016, O - C values are consistently between 40 and 60 seconds. Using mathematical analyses and existing models of the system, I concluded through my study that V808 Aurigae does have an exoplanet orbiting it. This research is important because it places our solar system in perspective, challenging current understandings of orbital mechanics and the diversity of solar system formation, and potentially advancing the search for extraterrestrial biology. Further research should be conducted on the system to increase confidence in my conclusion and uncover the properties of V808 Aurigae’s planet.
Elevator Pitch Video:
School: Angola High School
Grade: 10
Category: BMED
Abstract: Lacking knowledge about malignant neoplasms is a predominant issue for patients and people around the world. Without knowledge of what to look out for, people do not know when to seek medical attention because they are not aware that they have cancer. The methods and materials I used in gaining an in-depth understanding of any disease is the first step in knowing what to look out for, and that is what led me to choose this project. I began with the idea that cancer affects every family. I expanded that with the intention to spread awareness and provide a detailed explanation of malignant neoplasms. I started by researching each category individually, then cross-referencing my information and filtering through what I wanted and what I did not want to include. Finally, I gathered all of my online material and put it together. Cancer is one of the leading causes of death in the world, and being able to understand and comprehend it is a fundamental skill for raising awareness and recognizing risk factors. The future of medicine is ever-changing: new cancer studies and drug trials are coming out daily. The goals of this project are to raise awareness of the characteristics of malignant neoplasms and to promote caution regarding potential risk factors that can increase the likelihood of developing a malignant neoplasm now or in the future.
Elevator Pitch Video:
School: DeKalb Middle School
Grade: 7
Category: ROBO
Abstract: I Robot - Automated Color Sorting with FANUC Robot and Gray Scale Vision This project investigated the accuracy of a robotic system in sorting objects by color using a gray scale camera. The experiment looked to determine if robotic systems could pick and sort the test materials based on the vision system’s gray scale camera data. Hypothesis: The robot would achieve at least 70% accuracy in sorting brown and gray bricks randomly placed in the work area. System Setup: A fixed-mount camera was calibrated using a calibration grid and part height. Lighting was optimized using low-intensity LED and a red camera lens filter and to minimize glare. Programming: A robot program was developed to locate and pick the bricks using the vision system. The system also incorporated a Histogram tool to convert visual data into gray scale values (0-255) to distinguish between the two brick colors. Testing: Trials consisted of separate control groups for brown and gray bricks, followed by five mixed-color trials featuring 12 bricks each in random orientations. Initial Performance: Using an initial threshold of 125, the robot achieved a 90% overall accuracy in the mixed brick trials. It identified brown bricks with 100% accuracy but struggled with gray bricks (90% accuracy) due to the initial threshold. Optimization: Adjusting the gray scale threshold to 115 eliminated sorting errors. Conclusion: The final trials resulted in 100% accuracy across all groups. The results supported the hypothesis and demonstrated that while gray scale vision is highly effective for color sorting, precision tuning of histogram parameters is essential to overcome subtle lighting and material variations.
Elevator Pitch Video:
School: Ryan Park Elementary School
Grade: 5
Category: CHEM
Abstract: Objective: I chose this Project because I wanted to know why the cakes we enjoy are so delicious and fluffy. I also chose this project because I wanted to explore the effect that baking powder has on baked goods and how it changes their texture and taste. Research Questions: Does adding baking powder to cake batter cause a chemical reaction that helps the cake rise and become fluffy? Does adding more baking powder to cake batter make the cake rise even higher and become fluffier? Hypothesis: Adding baking powder to cake batter causes a chemical reaction that produces carbon dioxide gas, which gets trapped in the batter, resulting in the cake rising and becoming fluffy. Adding more baking powder to the cake batter will the cake rise higher but only up to a certain limit. After that limit, the baking powder will negatively affect the taste, height, and texture of the cake. Materials: Ingredients Control Trial 1 Trial 2 Trial 3 Trial 4 Baking Powder (Tbs) 0 ½ 1 2 4 Egg 3 3 3 3 3 Flour (Cup) 3 3 3 3 3 Vanilla powder (Tsp) ¾ ¾ ¾ ¾ ¾ Sugar (Cup) 1 1 1 1 1 Oil (Cup) 1 1 1 1 1 Milk (Cup) 1 1 1 1 1 Procedure: In my experiment, I used the same ingredients for all five trials, and the only variable was the amount of baking powder used in each trial. The amounts of baking powder were as follows: Experiment Control Trial 1 Trial 2 Trial 3 Trial 4 Baking Powder Zero ½ Tbs 1 Tbs 2 Tbs 4 Tbs Making the Cake: My mom preheated the oven to 350F. I mixed the dry ingredients in a bowl, whisked together the flour, baking powder, and salt. I set them aside. In a large bowl, I used a handheld mixer to mix the oil and sugar. I then added the eggs one at a time, mixing them well after each addition. I mixed the vanilla powder. I combined the dry and wet ingredients. I gradually added the dry ingredient mixture to the wet ingredient mixture, alternating it with milk until the batter was smooth and slightly thick. I poured 1 inch of batter into a 2 inch deep small, clear, glass bowl My mom baked the batter in the preheated oven for 25-30 minutes. My mom removed the cake from the oven and let them cool for about 10 minutes. I measured the height of each cake using a ruler and a measuring tape. I logged my data into Google Sheets and created a graph to show the effect of baking powder on cake making. Average Cake Heights: I conducted three experiments using the same ingredients and recorded the measurements for each trial, including the control. I calculated the average heights of the cakes from each trial across three experiments. The average heights were as follows: Batter Control Trial 1 Trial 2 Trial 3 Trial 4 Experiment 1 (Inch) 1 1.1 2.7 3.3 3.05 2.85 Experiment 2 (Inch) 1 1 2.75 3.20 3 2.75 Experiment 3 (Inch) 1 1.2 2.8 3.25 2.95 2.65 Average height (Inch) 1 1.1 2.75 3.25 3 2.75 Results (Height): The first thing that I measured was the height of the cake. I used a ruler and a measuring tape to measure how tall each baked cake was compared to the raw batter. I made sure to use the same size glass bowls and measuring cups for each. The table below shows the starting height of each batter, the height of each baked cake, and the difference in height between the batter and the finished cake. Control Trial 1 Trial 2 Trial 3 Trial 4 Height of batter (Inch) 1 1 1 1 1 Height of baked cake (Inch) 1.1 2.75 3.25 3 2.75 Difference in height (Inch) 0.1 1.75 2.25 2 1.75 Scatter Chart showing height of each baked cake: Column Graph showing height of each baked cake: Results (Fluffiness and Taste): The second thing that I observed was the fluffiness and taste of each trial cake, including the control cake. Control (No Baking Powder): The cake without any baking powder was compacted and it tasted awful. It was almost inedible. Trial 1 (½ Tbs of Baking Powder): This cake was sticky and fluffy at the same time. It was hard to swallow, but it tasted okay. Trial 2 (1 Tbs of Baking Powder): This cake was super fluffy, light, and soft. It tasted absolutely delicious. Trial 3 (2 Tbs of Baking Powder): This cake was still fluffy, but it had a bitter taste and was also a little sticky. Trial 4 (4 Tbs of Baking Powder): This cake was sticky and crumbly, and the middle was almost hollow. The top of the cake was burnt, and it tasted bitter with a weird metallic flavor. Overall, adding little or too much baking powder instead of the amount I used in Trial 2 made the cake taste and feel bad. This is because the baking powder reacts with the batter up to a certain limit. If we add too much baking powder, it can stay in the cake and make it taste bitter and sticky. If we add too little baking powder, the cake will not rise enough and it will become sticky and compact, making it dense and not fun to eat. Conclusion: In this experiment, I learned that baking powder causes a chemical reaction that releases carbon dioxide. This gas forms bubbles that get trapped in the batter, making it inflate and expand. This is what makes the cake fluffy and soft. I also discovered that increasing the amount of baking powder in the cake batter will help it rise higher, but only up to a certain limit. After that limit is reached, the cake will not rise anymore, and the surface may burn while the middle becomes crumbly instead of fluffy. In addition, I learned that adding more baking powder changes the taste of the cake. The more baking powder I added, the more metallic and bitter the cake tasted. I discovered that baking is not just an art, but also a science-based chemical process. This experiment was very beneficial for me because I learned how to use Google Sheets, I learned about the chemistry of carbon dioxide, and I learned how to calculate the average in math terms. Both parts of my hypothesis were supported by the results of these experiments. This science fair project shows how chemical reactions can affect everyday lives.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Marian High School
Grade: 11
Category: BCHM
Abstract: Gold nanoparticles have a wide range of applications and on a small scale have many electrons available to react making them convenient to attach other molecules to through the use of thiols. The goal of my project was to synthesize gold nanoparticles and measure the binding of Immunoglobulin G and gold nanoparticles to protein A or G using UV-Vis spectrophotography. My methods were synthesizing the gold nanoparticles by preparing a gold citrate solution and heating it, conjugating Immunoglobulin G to nanoparticles, adding protein G to the immunoglobulin G bound nanoparticles. attempting to calculate the aggregation index, and run the solutions on a UV-Vis machine. I calculated the diameter, number of gold nanoparticles, the surface area per particle. the amount of Immunoglobulin G, the loading of Immunoglobulin G per molecule, the theoretical size and loading, and the estimated antigen binding capacity. There was a high loading of Immunoglobulin G on each nanoparticle, so I was unable to calculate the aggregation and binding constant of Protein G. I was successfully able to quantify the loading of Immunoglobulin G per nanoparticle and it aligned with what was theoretically expected. This project has many real world applications such a as point of care diagnostics, immunotherapy protocols, and protein purification processes. In order to calculate the aggregation I would repeat this protocol with a lower concentration of Immunoglobulin G to more easily allow the proteins to bridge between the nanoparticles and aggregate.
Elevator Pitch Video:
School: Marian High School
Grade: 10
Category: BCHM
Abstract: When food is cooked and processed in any way, it risks losing some of its nutrients to water and high temperatures. This study seeks to investigate which standard method of cooking keeps intact the greatest amount of the vitamin b12 (cobalamin) in nori, a b12-rich dried seaweed made of algae that serves as a cobalamin source for individuals without animal products in the diet. This experiment used Ocean’s Halo Sushi Nori. The control sample was one sheet left uncooked. The cooking methods used on the remaining three samples were as follows: boiled in water for five minutes, microwaved for 90 seconds, and oven-baked at 400 degrees Celsius for ten minutes. The resulting samples were then each emulsified with 250mLs water to form a solution, and each solution was scanned using a UV-Vis spectrophotometer at wavelength 360nm to detect the remaining concentration of b12 (cobalamin). The results supported that the uncooked nori had the highest presentation of b12 at a mean concentration of 1.055, while the microwaved nori had the highest remaining b12 levels of the cooked samples with a mean concentration of 0.934. The baked nori presented a mean concentration of 0.607. The boiled nori presented a mean concentration of 0.268. The highest value of the vitamin was present when no cooking at all occurred. Of every cooking method, microwaving was the least detrimental to the cobalamin content.
Elevator Pitch Video:
School: Marian High School
Grade: 11
Category: BMED
Abstract: This experiment tests how ACE2 genotypes affect peoples' various symptoms of COVID-19.I wanted to test this because everyone has the ACE gene which allows the ACE2 enzyme to help regulate ANGII. There is an ACE gene variant called the I/D (Insertion/Deletion) variant which produces 3 genotypes: II, DI, and DD. These differ among individuals because each genotype influences how much of the ACE enzyme the body produces. ACE2 acts as a receptor for COVID as the virus is a spiked protein and it latches onto ACE2. This attachment initiates entry into the cells to become infected with COVID and when it's bound to the ACE2 receptor, this allows ANGII to injure tissue and contribute to the lung and heart injuries that COVID-19 patients usually have. Not everyone experiences the same symptoms though. My hypothesis is if we are able to test peoples' DNA, who’ve had COVID-19 before, then we’ll see a pattern with how their symptoms affected them corresponding to their ACE2 genotypes.
Elevator Pitch Video:
School: East Noble High School
Grade: 12
Category: BCHM
Abstract: In order to determine whether Bisphenol-A (BPA) exposure can cause biological changes, this project examined how BPA impacted Drosophila melanogaster reproductive capacity and behavior. Flies were exposed to three types of bisphenols: Pure Bisphenol-A (PBPA), Thermal receipt paper containing BPA (RBPA), and BPA-free thermal receipt paper (RBPS). Reproductive capacity was then measured at three different time periods via male, female, and total populations. Behavior was tested via pupal deposit height, the lowest, highest, and range, as well as an examination of larval activity. The data was then analyzed utilizing chi-squared tests, comparing each treatment group to the control. In terms of reproductive capacity, significant results appeared across all categories. PBPA appeared as significant across each time period and gender. RBPA showed significance in male and total counts, but appeared to have no effect on female population. Finally, RBPS had a varied significance pattern that showed no direct trend. In terms of behavioral significance, there was little to none. The only instance of significance appeared in a PBPA larval activity sample. Based on the chi-squared values proximity to a .05 p-value, there appeared to be very little correlation between the results and the variable. These results suggest that exposure to BPA can alter reproductive capacity in Drosophila melanogaster, but the extent varies based on concentration and method of exposure.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Eastern Hancock Middle School
Grade: 6
Abstract: This student has not yet submitted an abstract.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Westview Elementary School
Grade: 6
Category: BEHA
Abstract: We did this project because one of us wants to be a psychologist when she grows up due to her interest in the brain and how it works. Both of them want to have future reference for when they have to read material for school,that way they know the better way to read for maximum comprehension of the text. To complete this project, first we would take our test subject and sit them down in the quiet hallway. Our hypothesis was that people would remember a written piece better when reading it off of print rather than a screen. Due to the fact that when you read off print you are engaging more senses like vision, smell, and touch. By engaging these senses you are more likely to focus and remember the text better. For this project we needed several materials. The materials we needed were: 10 female test subjects, 4 poems by Shel Silverstein, 1 chromebook, 1 printer, 1 number 2 pencil, 1 stopwatch, 1 notebook, 1 quiet hallway, and 40 lined pieces of paper 11 inches by 8 ½ inches. To complete this project, first we would take our test subject and sit them down in the quiet hallway. Next, one of the conductors would give them the instructions and the rundown of what was about to happen. While this was going on the other conductor would set the stopwatch to one minute. Then we would show them a poem either on the chromebook screen or on paper depending on the trial, and start the timer. Once the timer was up we would take away the poem. Then we would give the subject a clipboard with blank, lined paper and a number two pencil. Also at this point we would tell the subject “to write the poem down word for word as best as they can”. Next we would give them time to follow the instructions. After we would send the subject away or back to their classroom. Once the subject was gone, one of the conductors would compare the subject’s written response to the actual poem. We would count them off if they left words out or switched words around. Once we had counted off all the mistakes, we would turn the fraction into percents.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Carlin Park Elementary School
Grade: 4
Category: CHEM
Abstract: The purpose of this experiment was to determine which potion I could find around the house that cleans tarnished pennies most effectively. I researched this and found out that vinegar with salt should clean pennies best because it is acidic and can dissolve copper oxide which is what makes the pennies look brown. To test this, I soaked 25 total pennies in five different potions. This was five trials with five pennies in each trial. My five potions were Essence of Erosion (vinegar with salt), Sour Serum (lemon juice), Night Bubble Brew (RC Cola), Aqua Purificata (Dawn dish soap & and water), and Phoenix Ember Elixir (ketchup) and I soaked the pennies in them for five minutes. After soaking, I rinsed and dried the pennies and compared their brightness. The results showed that Essence of Erosion (vinegar with salt) cleaned the pennies most effectively. This experiment demonstrates how acids react with copper oxide to remove tarnish.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Butler Elementary School
Grade: 5
Category: EAEV
Abstract: It’s important for people to know about my project because it’s so they know about the different stomata density on sun vs. shade because what if someone doesn’t know about the stomata density on plants leaves.In my procedure so first I labeled 20 cups 10 with shade and 10 with sun,then I put ½ in each cup, next I put 2 seeds in all the cups ¼ deep in each cup,then after that I gave each plant 1 tbsp, next I put 10 cups into the shade and 10 cups into the sun,then repeat giving it ¼ of water for 6 weeks then when I was done I record and measure the stomata density on each plant. The data I collected is all about the stomata density on plants leaves in different locations; some plants grew in the sun and some grew in the shade. My results for the sun was 6 as the average and the shade had 4 as its average. I was correct and my results show that I was correct shade and sun do affect the amount of stomata density sun had more stomata density than shade.I met where my design was I like to think that I could have done more but I did what I could do from it but if I had to do it again I definitely start sooner, water it more, and for sure do a different type of plant.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Terre Haute South Vigo High School
Grade: 9
Category: EAEV
Abstract: In this project, I will be creating a precipitation prediction model for the Sub-Saharan Region (from roughly the north of Zambia to the south of Botswana). This region is very vulnerable to variations in precipitation levels, as a large percentage of the population practices subsistence farming. The features for this model include a dendrochronological precipitation reconstruction, and satellite data from the region. To create the dendrochronological reconstruction, I sampled in Zambia, and combined this chronology with 4 others from the Sub-Saharan region. I downloaded the satellite from NASA's Earthdata website. After preprocessing all of these features, I put them into a mix of gradient tree models and neural networks. I then stacked the results of these models to create an ensemble prediction model for the Sub-Saharan region.
Elevator Pitch Video:
School: Penn High School
Grade: 12
Category: BCHM
Abstract: Antibiotic resistance, the process by which a bacteria adapts to be able to survive antibiotic treatment, is of increasing concern. Liposomal enclosure of existing antibiotics has appeared as a popular method of circumventing antibiotic resistance, being shown to improve antibiotic efficacy in resistant strains. However, limited research has examined its effect on non-resistant strains. Solutions of 10, 50, and 100 µg/mL of ampicillin and streptomycin were prepared, along with equivalent solutions containing liposomes. These liposomes were made from injecting water onto dry phospholipids. We scratched E. coli onto 1 cm2 square segments on a petri dish, then tested the inhibitory effects of the solutions on the non-resistant E. coli samples, measuring the diameter of the area of inhibition on the bacteria. A two-way-ANOVA test was performed to determine the significance of the inhibitory effect of either antibiotic with versus without liposomal enclosure at the aforementioned concentrations. The p-values for liposomal versus non-liposomal treatments were 0.112 for ampicillin and 0.270 for streptomycin (α = 0.05), indicating no statistically significant difference. The lack of significant difference may have been due to improper liposomal enclosure. Alternative preparation methods, such as the ethanol injection method, should be tested to determine if the insignificance was due to improper enclosure. Additionally, since E. coli is Gram-negative, limited interaction between liposomes and the outer membrane may have reduced effectiveness. Using the same or similar procedure on a gram-positive bacteria should be performed to determine whether cell wall structure influences liposomal antibiotic delivery.
Elevator Pitch Video:
School: NIRSEF
Grade: 7
Category: BEHA
Abstract: AI images have become very realistic, so can people identify which pictures are real, and which age range would do the best? This project looks at how different age groups, 12-24, 25-45, 46-60, and 60 and older perform when identifying AI images. Using AI photos from different generators and real photos, I tested the average correct answers each group would make. My hypothesis was that on average, people could not identify AI from real photos correctly. I also said that the 12-24 age group would do the best since they have grown up in a technological age. This experiment disproved my hypothesis by showing that 53% of the responses were correct. Also, of all the age groups the 46 to 60 year olds did the best, disproving my hypothesis. Although, the results were so close it would make sense to do further research into this project to see if there is a difference when more people participate.
Elevator Pitch Video:
School: Saint Mary Elementary School
Grade: 8
Category: CHEM
Abstract: My project was testing which type of milk can make the strongest glue. I hypothesized that skim milk would create the strongest glue because it has the least amount of fat, resulting in stronger glue. To test this, I started by creating the glue. To make the glue, I warmed 2 cups of 2% milk to 100 degrees Fahrenheit. Next, I added 3 tablespoons of vinegar to separate the whey from the casein. Then I strained the whey and casein. I then added the casein back to the pot and added 1 tablespoon of baking soda to deacidify the casein and 4 tablespoons of water to create a glue-like substance. I then did the same process for whole milk, skim milk, and oat milk. To test my glue, I added the homemade glue to 2 pieces of cardboard together and let them dry overnight. Finally, I hung a 1 inch spring clamp to the bottom of the cardboard and then hung a gallon bucket from the clamp. To test my hypothesis, I kept adding water into the bucket by cups until the bucket was full. I discovered that the whole milk, skim milk, and 2% reduced fat milk can hold up to at least 1.25 gallons of water. On the other hand, oat milk cannot make glue due to the lack of casein in it.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Lafayette Tecumseh Jr High School
Grade: 7
Category: ROBO
Abstract: My project is an AI using the NEAT algorithm attempting to climb as high as it can in a platformer game using different vision methods. I chose this project to build on a previous experiment I conducted, which was AI solving a maze. My hypothesis was that keypoint vision, one of my vision types, would be able to climb the highest. So, for my experiment, first, I had to code the program and set up the game. Next, I put an AI in and ran it for 100 generations. Then, I put its highest height climbed to in my data table and switched to the next AI. However, opposing my hypothesis, the control (no vision) performed the best. I learned from this experiment how NEAT works and how to better code in Unity (the software I was using).
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: St John Lutheran School
Grade: 6
Category: PLNT
Abstract: Does storing bananas in different ways cause them to stay yellow longer? I tested whether storing bananas with their stems wrapped, hanging them up, or putting them in green bags helped them stay yellow the longest. I predicted that wrapping the bananas’ stems would enable them to have the highest percentage of yellow, as it may block the ethylene from reaching the banana more than the other variables. I ran two experiments with sixteen bananas each. There were four bananas in each group: stems wrapped, hanging, green bags, and control. I took pictures of the bananas before and after the experiment, then analyzed them using Fiji, a computer software that measured the percentages of yellow (white) to the amount of brown (black). My hypothesis was that the stems-wrapped bananas would have the highest percentage of yellow after 10 days. The data did not support my hypothesis. The average amount of yellow for the stems-wrapped bananas changed from 49-56% in both groups. This was hardly better than the control group, which changed from 56-65%. The hanging bananas did slightly better; they changed 41-52%. Surprisingly, the green bag bananas changed the least both times, only 2-6% from the beginning to the end. I inferred that the green bags contained a substance that was very effective at absorbing the bananas’ ethylene. Wrapping the stems and hanging the bananas did not stop ethylene’s basic effects. While keeping bananas from turning brown may be challenging, these results can help families enjoy their yellow bananas longer.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: NIRSEF
Grade: 8
Category: ENBM
Abstract: Children with Sensory Processing Disorder struggle to keep calm and focused in a school environment. Previous solutions for this problem such as a weighted lap pad do not have textures or interactive items to help children. In this prototype it would need to be 7% of the targeted user's body weight. The targeted user was a 5 year old and their approximate body weight is 40lbs making the prototype nearly 3lbs. Various textures must also be included. After building and observing it was concluded that the child would be able to find the binder useful to solve their problem. This is because the binder met the requirements of weight and interactive textures which were crucial for this problem. The child would also be able to use the binder in the way it was intended seeing that it is easily held and movable.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Sarah Scott Middle School
Grade: 8
Abstract: This student has not yet submitted an abstract.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Saint Mary Elementary School
Grade: 6
Category: CHEM
Abstract: The purpose of this experiment was to determine which type of wood burns the longest and is best for heating. Different types of wood are commonly used as fuel. Many kinds of wood vary density and moisture content which may affect how long they burn.The types of wood tested in this experience were oak, yellow pine, popular, ash, construction pine and western cedar. Each wood sample was cut the same size and weight to ensure a fair contest. The samples were burned individually under the same environmental conditions. A timer was used to measure how long each of the pieces burned from ignition until the flame completely stopped. The results showed that oak burned the longest, followed by pine and popular. Oak is hardwood and is denser and burns more slowly. In conclusion, the experiment demonstrated that wood type effects burn time,and hardwoods such as oak are more efficient for longer-lasting fires. Hardwood burns longer than soft woods making Oak a good choice for our heating and camping needs.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Saint Mary Elementary School
Grade: 7
Category: ETSD
Abstract: This student has not yet submitted an abstract.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: West Lafayette Intermediate School
Grade: 6
Category: PLNT
Abstract: This science experiment proves that different types of fertilizer affect how plants grow and affect soil health. I tested what fertilizer (worm castings, manure, miracle gro) is most beneficial to soil and plant growth. For a short summary of what I did I put different amounts of the three fertilizers in pots that have the same amount and type of soil, and watered them everyday. After four weeks I brought them to the Purdue Lab to run tests such as CO2 Respiration levels and Root Mass levels. After I tested them I found out that biofertilizers such as manure and worm castings had better results than the other fertilizers. This experiment can help many agricultural workers around the world when they need to choose a fertilizer to grow their crop and plants.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: West Lafayette Intermediate School
Grade: 6
Category: BMED
Abstract: I have always had chapped lips and was curious which one was the most sterile and which one had the most benefits. So I decided to test five different brands of lip balm to see which one has the most benefits. I tested the Sun Protectant Factor, moisture barrier, and if any of them grew bacteria. I tested if any of them grew bacteria by swabbing each lip balm brand new, used on silicone tape, and used once on my lip. I tested the moisture barrier by putting fabric on a tube and putting lip balm on the fabric.Then after three days I measured the distance from the beginning water level and the current water level. I tested the SPF effectiveness by putting beads in a 3D printed dish and putting a petri dish lid covered in the lip balm and taking it outside. Then I had some of my neighbors rate each one out of five, one being the lightest and five being the darkest.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: West Lafayette Intermediate School
Grade: 6
Category: BEHA
Abstract: This study examined whether different types of background sound affect verbal memory recall. This experiment is important because it will show if using a certain background noise can affect your studying productivity and/or accuracy. Forty-eight participants completed a word recall task under four conditions: instrumental music, music with lyrics, white noise, and silence. Participants were given 12 words in each condition and had 45 seconds to recall as many words as possible. After each condition, they rated their level of distraction on a scale from 1 (not distracted) to 5 (very distracted). Paired t-tests showed no statistically significant differences in recall between sound conditions. However, in the lyrics condition, distraction ratings were negatively correlated with recall (R² = 0.20), suggesting that participants who felt more distracted remembered fewer words. These findings suggest that individual differences in perceived distraction may influence memory performance more than sound type alone.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: West Lafayette Intermediate School
Grade: 6
Category: PHYS
Abstract: My science fair project is to see what I can do to keep a rubber band helicopter flying for the longest time. I attached the rubber band(s) to the plastic hook and the hook on the propeller on the 1.5”7 helicopter. I tested two variables: the first was the twisting amount of the rubber band, 20, 24, 28, 32, and 36 times, and recorded the times. The second variable was how many rubber bands I used, 1, 2, and 3 rubber bands, and recorded that time. The result was what I hypothesized, as the more I twisted and added rubber bands, the time of flight increased.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: West Lafayette Intermediate School
Grade: 6
Category: CHEM
Abstract: This student has not yet submitted an abstract.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: NIRSEF
Grade: 4
Category: ENEV
Abstract: This student has not yet submitted an abstract.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: NIRSEF
Grade: 6
Category: PHYS
Abstract: I started out doing a cardboard car and seeing the different distances it would track with weight but the wheels would not move or turn also the balloon would not propel it forward, so due to the failed test I tried something more impact without wheels. I used gravity with magnets as my weight and subject of my experiment, and the car floated but took wasn’t as good as I wanted my experiment. So now I am doing a new adaptation by different materials, that will make my car more stable when it floats.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: NIRSEF
Grade: 5
Category: EAEV
Abstract: Macroinvertebrates can tell us about how healthy the water is in a stream. Macroinvertebrate numbers can change in the Fall due to piles of leaves gathering in streams. It was hypothesized that macroinvertebrate numbers would be low in the Fall because leaf litter piles can cause low oxygen. Dissolved oxygen readings were taken and macroinvertebrates were identified at 6 locations in Northwest Indiana. The dissolved oxygen was below normal at 2 sites and all of the pollution readings using macroinvertebrate data was either poor or fair. The data shows the water may have been polluted at some sites, but intolerant macroinvertebrates, like mayfly nyphs and riffle beetles, were present at all 6 sites indicating a healthy ecosystem. Dissolved oxygen and macroinvertebrate populations can change naturally in the Fall when there is a change in seasons. Leaf litter piles can also disrupt the flow of the stream and that may also alter dissolved oxygen and the abundance of macroinvertebrates.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: NIRSEF
Grade: 5
Category: BCHM
Abstract: Bioluminescent dinoflagellates are tiny ocean organisms that produce light when they are disturbed. I first became interested in them after visiting a bioluminescent bay in Puerto Rico and seeing the water glow around our kayak. This made me wonder if water temperature affects how brightly these algae glow. My hypothesis was that if the water temperature increased, then the algae would glow more brightly. To test this idea, I placed nine identical vials of dinoflagellates into three temperature groups. The control group was kept at 21°C, the warm group at 23°C, and the hot group at 26°C. Each group contained three vials. I measured the brightness of the glow once each night at 8:00 PM for one week using a digital light meter. A metronome was used to make sure each measurement lasted the same amount of time so the data would be consistent. At first, the brightness of the glow was similar in all three groups. After several days, the algae in the hottest water began to glow brighter than the other groups. These results suggest that warmer temperatures can increase bioluminescence over time. Bioluminescence happens when a chemical called luciferin reacts with oxygen with the help of an enzyme called luciferase. Warmer temperatures may allow this reaction to happen more often. This experiment helps show how temperature can affect living organisms and may help scientists better understand how warming oceans could impact bioluminescent ecosystems.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: NIRSEF
Grade: 10
Category: EBED
Abstract: Throughout critical operations and natural disasters, communication is of the utmost importance. During such operations, centralized communication systems are often used, relying heavily on centralized infrastructure such as base stations or cell towers. This system introduces a single point of failure, which is extremely volatile during emergencies. This project investigates the development of a portable mesh Mobile Ad Hoc Network (MANET) system capable of operating in lieu of fixed infrastructure. The system contains nodes that have a microcontroller, radio frequency communication module, integrated GPS, and custom-made printed circuit boards. When the nodes are initialized, the nodes establish an adaptive mesh network, allowing for packets to be relayed between nodes using multiple hops without needing line-of-sight (LOS) radio communication. The system, Stress-Induced Link Evaluation for Network Tactical meshes (SILENT-MESH), integrates end-to-end encryption, Morse code fallback, GPS broadcasting, and distress signaling protocols. SILENT-MESH also utilizes radio frequency hopping to reduce foreign interference and improve connectivity in degraded and congested environments. SILENT-MESH and hardware testing demonstrate the network's ability to maintain and preserve accurate and efficient packet transmission despite node failure, demonstrating the resilient nature of a decentralized communication network. The results yielded from the project suggest that mesh communication systems can become a practical alternative for tactical field operations, emergency response teams, and remote fieldwork where typical fixed infrastructure is absent or inoperative.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: NIRSEF
Grade: 5
Category: PHYS
Abstract: Purpose: The purpose of this experiment was to investigate how different wheel materials affect the speed of a DC motorized toy car. When different materials contact a surface, they experience different amounts of rolling friction force. I wanted to test whether wheel material could change how fast a motorized toy car travels on a wood track. Methods: Three identical toy cars were built using the same design. The wheel diameter, rims, and axle length, were kept the same. The only difference between the cars were the wheel materials: steel, wood, or rubber. A straight track was built using a wooden board with guard rails. A starting line was marked to ensure four, consistent trials. Each car was raced individually, a stopwatch was used to record the time it took each car to reach the finish line. Results: The steel wheeled car was the fastest in all four trials, reaching the finish line in about 0.62 seconds. The wood wheeled car finished second, and the rubber wheeled car was the slowest at about 1.01 seconds. The rubber wheeled car took longer to pick up speed after it was dropped on the track. Conclusion: The results prove that wheel material does affect the speed of a motorized toy car. Materials that have lower rolling friction will allow the toy car to travel faster. This can be explained by the coefficient of rolling friction of each material against wood. The results can also be explained by the deformation of the materials on a hard surface. Steel wheels rolled the fastest because it deforms less than wood and rubber wheels. The steel wheels retained the most kinetic energy and was able to use it to travel faster. Real-World Application: Understanding rolling friction can help engineers design faster and more efficient vehicles, bicycles, and machines.
Elevator Pitch Video:
School: NIRSEF
Grade: 7
Category: MATS
Abstract: “Why is my clay crumbling up?”, “Why does it feel like the clay is not baked?”; these are questions that are asked when not knowing the correct temperature to bake polymer clay. As a clay artist, it is important to know what the best temperature for baking clay is, so you don’t ruin your masterpieces. This experiment examines what the optimal temperature to bake the polymer clay is to achieve the maximum durability. The clay was tested by counting the number of folds for each piece of clay until split into two pieces (dependent variable) after baking in an oven at different temperatures in Celsius that changed after each trial; 96℃, 113℃, 135℃ (independent variable) with the same amount (1.4 g) and shape (3 cm by 1 cm by 2 mm) of clay being used, the time (15 minutes) that the clay is baked, and the oven that is being used (controlled variable). The hypothesis was that if pieces of clay are baked for the same amount of time, but at different temperatures, then the piece of clay that had a higher temperature will have more durability. The results of this experiment supported the hypothesis that the pieces of clay that were baked at a higher temperature will have the most durability.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: NIRSEF
Grade: 8
Category: BEHA
Abstract: This experiment is important because it shows that eye movement will not give away subterfuge. This information helps people with their jobs and everyday lives. For example, police officers: if they are asking someone questions, studying the person’s eyes is not a reliable way to tell if that person is deceiving them. How does lying affect the eyes? I thought telling a lie would cause a person’s eyes to dilate, or look up and to the left. That theory was proven wrong by my experiment. A person's eyes will not do any one predicted thing. For my procedure, I asked 25 people six questions each. I told each person to lie on three of those questions and tell the truth on the others. On each question, I observed the movement and dilation of their eyes. I found that my hypothesis was wrong after collecting my data. Only 12% of people showed eye indications of subterfuge. In conclusion, lying does not affect the eyes in any one specific way. The results highly contradict my hypothesis because I thought that when someone is lying, their eyes will go up and to the left or dilate, but my results tell me that a person’s eyes could do absolutely anything when they lie. I would improve this experiment by testing more people so that the results are more accurate. I would also video record subjects while answering, so I could have the option to watch the recordings and make sure my markings are correct.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: NIRSEF
Grade: 6
Category: MATS
Abstract: My Project is about Eco-friendly paints. Most people don't notice it but, a lot of people use paints that hurt the environment. So I made four paints each with different ingredients. One, (Paint A) Is made with the original homemade paint with the main ingredient being flour. Paint B Is made with green grapes. Paint C is made with maple syrup. And paint D is made with shaving cream. I have organized and taken notes from the past few days on how they were each day. They had better quality as they sat in containers wrapped in plastic wrap. I want to remind everyone that saving the environment is one of the top priorities and must be taken in special care. For an example; Acrylic, craft paint, window glass paint, spray paint, and another type of acrylic paint are very bad. The worst one is spray paint with 95% toxic chemicals inside. And people like to use that outside. This is why I made four safer paints that are harmless for the environment. I also compared the ingredients, and every paint has to have the binder, the solvent, the preservatives and stabilizers. The binder is the thickness, the solvent is the liquid, and the preservatives and stabilizers are the things that don't cause algae. The flour is the binder, the water is the solvent, and the preservatives and stabilizers are the vinegar and salt. I really hope people start to realize and use safer paints for the environment.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: NIRSEF
Grade: 7
Category: EAEV
Abstract: Global warming is thawing permafrost (carbon-rich frozen ground covering 15% of the Northern Hemisphere). To understand how this thawing could impact ecosystems and climate change, a two-phase project investigated how freeze-thaw cycles (FTCs) (1) affect soil respiration/moisture/pH levels and (2) the amount/timing of algae growth. Phase 1: soil samples were exposed to 0,1,3,5,7 freeze–thaw cycles and CO₂ emissions measured over 7 days. FTCs led to higher soil respiration. The first FTC has the greatest impact as CO2 increased dramatically from 0 to 1 FTC. Freezing breaks open soil aggregates and microbial cells, releasing carbon/nutrients which increases microbial activity causing a burst of CO2 emissions. Additional FTC resulted in lower pH and higher moisture levels driving a longer period of increased soil respiration. Phase 2: water was added to soil, and supernatant was decanted and exposed to 0,1,3,5,7 FTCs before being added to algae samples. Algae growth was measured over 18 days using a spectrophotometer to test absorption at optical density 750 nm. Over 13 days, all samples exposed to FTCs had less and slower algae growth. FTCs led to microbial death, reducing the size of the microbial biomass available to respire nutrients that drives algae growth. By day 14, all samples converged to similar absorption levels as nutrients were consumed. FTCs increased CO2 emissions but did not increase algae growth. Implications: as permafrost thaws, the Arctic could change from a beneficial carbon sink to a contributor of global warming; delays in algae growth could cause unknown impact to ecosystems.
Elevator Pitch Video:
School: NIRSEF
Grade: 8
Category: CHEM
Abstract: This student has not yet submitted an abstract.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.
School: Culver Academies
Grade: 10
Abstract: This research examines how exposure to emotional content on social media, beyond time spent online alone, influences young adult mental health, with a focus on sleep as a behavioral pathway and content-aware feedback as a potential intervention. Using a multi-study design integrating behavioral science, statistics, and data science, four studies were conducted. Study 1 analyzed survey data from 503 young adults (ages 18–25) and found that exposure to negative emotional content was significantly associated with higher depressive symptoms, with sleep quality mediating this relationship. Study 2 used an experimental design in which participants were randomly assigned to view social media posts varying in emotional tone; exposure to negative content led to worse immediate mood and greater intentions to delay sleep. Studies 3 and 4 introduced and evaluated Mooditer, a privacy-preserving application that analyzes and summarizes users’ emotional exposure patterns. Compared with time-only feedback, content-aware feedback provided by Mooditer significantly increased awareness of emotionally risky exposure and strengthened intentions to regulate nighttime social media use. Together, these findings suggest that emotional characteristics of social media content play a critical role in young adults’ well-being and highlight the potential of targeted feedback tools to support healthier digital behavior beyond conventional screen-time limits.
Elevator Pitch Video: This student has not yet submitted an elevator pitch video.