2021 Senior Flint Regional Science and Engineering Fair
Flint Regional Science and Engineering Fair
Public viewing for the 2021 Senior Division of the Flint Regional Science and Engineering Fair.
Projects are conducted by 9-12 grade students across mid-Michigan.
Learn more about the FRSEF at www.flintsciencefair.org
More info: https://flintsciencefair.org/2021Senior
Filter displayed posters (110 keywords)
COVID-19 (2)
Chemistry (2)
Medicine (2)
computer science (2)
show more...
3D Printing (1)
A.I. (1)
AI (1)
Absorbance (1)
Additive Manufacturing (1)
Aerobic exercise (1)
Aging (1)
Analysis (1)
Animal Sciences (1)
Antibiotic resistance (1)
Arduino (1)
Arrhenius Equation (1)
Biology (1)
Biomedical (1)
Botany (1)
C. elegans (1)
Cancer (1)
Chemical Kinetics (1)
Cocrystal (1)
Coffee (1)
Cold Chain (1)
Congestion (1)
Convolutional Neural Network (1)
Diesel Exhaust Fluid (1)
Efficiency (1)
Environmental Science (1)
Genomics (1)
Health (1)
Hospitals (1)
Hot water (1)
Hydrogels (1)
Invertase (1)
Iodine Clock Reaction (1)
Kinetic Degradation (1)
Lifestyle (1)
Lightweight (1)
Low cost (1)
Lung cancer (1)
MIC method (1)
Machine learning (1)
Materials Science (1)
Mean Kinetic Temperature (1)
Mental Health (1)
Mushrooms (1)
Mycelium (1)
N.E.A.T. (1)
NEAT (1)
Newton (1)
Peltier modules (1)
Personalized Medicine (1)
Phytoremediation (1)
Plant Science (1)
Plants (1)
Potency (1)
Radiographs (1)
Radiomics (1)
Rapid detection (1)
Reaction Rates (1)
Recoil (1)
Regeneration (1)
Resource conservation (1)
Selenium (1)
Smartphone capable (1)
Streptococcus mutans (1)
Sunflowers (1)
Tool (1)
Toothbrush (1)
Urea Percentage (1)
Vaccines (1)
Vitamin C (1)
Water Filtration (1)
X-ray (1)
antibacterial effect (1)
artificial intelligence (1)
artificial neural networks (1)
backpropagation (1)
bacteria (1)
beverages (1)
bike (1)
coding (1)
computer art (1)
curcmin (1)
deep learning (1)
diet (1)
elections (1)
essential oils (1)
genetic algorithm (1)
gradient descent (1)
health (1)
machine learning (1)
mathematics (1)
mouthwash. (1)
neural networks (1)
painting (1)
programming (1)
psi (1)
reinforcement learning (1)
reproduction (1)
spectrofluorometer (1)
spectrophotometer (1)
spring scale (1)
statistical analysis (1)
supervised learning (1)
unsupervised learning (1)
weight (1)
β-Lactoglobulin (1)
Show Posters:
Tracks
▼ 4th Award Back to top
Smartphone Capable Lightweight Convolutional Neural Network Model for Detecting COVID-19 in Chest X-rays: Addressing the need of Resource-strapped Locations
Ayan Nair (Mentors: Professor Adam Alessio & Wenjie Qi, Michigan State University)
Abstract
COVID-19 has been a global health crisis for the past year, and scientists are still finding ways of preventing the spread of the virus as well as detecting it. One way of testing COVID-19 is with the use of chest radiographs. You take an x-ray, give it to a radiologist, and get your results within 24 hours. However, many hospitals do not have a radiologist on hand, making it harder for hospitals to give accurate and timely results to patients. An article from The Atlantic stated that “There are more radiologists working in the four teaching hospitals on Longwood Avenue in Boston, Massachusetts, than there are in West Africa,” with about 130 radiologists in Massachusetts General, Boston, and only 2 working in the country of Liberia. These statistics necessitates an alternative to a radiologist in these resource-strapped locations.
Accounting for the rapid rise of deep learning throughout the last decade, this project aims to use a lightweight convolutional neural network to accurately detect COVID-19 from standard chest radiographs. Specifically, a lightweight neural network that can be used on a smartphone, which allows it to be mobile and to be used in locations where a radiologist is not available. My project used two convolutional neural networks; a ResNet model and a lightweight mobile neural network. A ResNet is a fully convolutional neural network that can identify and classify an input image into 1000 object categories. These models are heavy in computational power and are very “deep” (contain a lot of layers). The lightweight mobile neural network, on the other hand, is a highly stripped-down version containing less than 15 layers. The research question my project aims to answer is if a lightweight neural network tool can accurately detect COVID-19 pathologies from chest radiographs. To answer the research question, I first gathered 1345 viral pneumonia chest radiographs provided by Chowdhury et. al. (2020), Rahman et. al. (2020), 1070 images of COVID-19 radiographs from GitHub provided by Cohen et. Al. (2020), and 1358 images of normal chest radiographs obtained from CheXpert (Rajpurkar et. al. 2019), a large public radiograph dataset. Using 3000 chest radiographs from this data, I first trained the ResNet model and used the weights from the training model to test 800 radiographs. The performance of this model (accuracy, sensitivity, specificity) was compared with the performance attained from the lightweight model. The results show that the lightweight model provided very high accuracies, sensitivities, and specificities, which are 98.4%, 97.7%, and 98.7% respectively. This offers potential for using such a smart phone capable lightweight model in resource strapped locations.
Presented by
Ayan
Institution
Okemos High School
Keywords
Convolutional Neural Network, Lightweight, Smartphone capable, Tool, Low cost, X-ray, COVID-19
▼ 5th Award Back to top
Radiomic Classification of Chest Radiographs - Identifying COVID-19, Viral Pneumonia, and Healthy Patients to Alleviate Hospital Congestion
Arohi Nair (Mentors: Prof. Adam Alessio and Wenjie Qi, Michigan State University)
Abstract
With the surge of COVID-19 over the past year, the need for fast and reliable detectors of the virus has reached a critical level. This Random Forest based radiomic classifier model works to quickly and effectively detect COVID-19 through the classification of chest radiographs. With fast methods such as this one, hospitals may be able to limit the overcrowding and backlog they have been experiencing by prioritizing high-risk cases. Radiomics could rapidly triage patients, causing resources to be conserved and allowing for hospitals to work more efficiently to help people who are afflicted with COVID-19.
Radiomic analysis is a novel medical imaging technique that has displayed high accuracy performances while also having a quick runtime. This method involves extracting radiomic features from clinical images, and inputting those values into a classifier model. In this study, I use radiomics to classify pathologies in chest radiographs from an open source dataset by Chowdhury, et al.; Rahman et al. (2020). I start with a set of 96 radiomic features and reduce it down to a list of 10 definitive features. These radiomic features almost perfectly differentiate between radiographs with COVID-19, viral pneumonia, and no pathology, with an accuracy of 97.0%. This reduced model can help hospitals with overcrowding caused by COVID-19 by using these specific features in radiomic analysis to determine which patients have COVID-19 and are at the highest risk. At the moment, detection methods of COVID-19 such as PCR tests, and even the rapid tests take a significant amount of time for the results to be returned. In fact, PCR tests take 24 hours at the minimum, with rapid, take-home tests just slightly shorter. Additionally, sending chest radiographs to radiologists for analysis takes time, and also places excess stress on the radiologists. In comparison, radiomics takes a matter of seconds and is a completely automated process. Current technology relies on time-consuming methods such as the bacteria growth in conventional COVID-19 tests, and the human examination that radiologists carry out; in this study, radiomics is examined as a functional alternative.
Radiomic analysis is a novel medical imaging technique that has displayed high accuracy performances while also having a quick runtime. This method involves extracting radiomic features from clinical images, and inputting those values into a classifier model. In this study, I use radiomics to classify pathologies in chest radiographs from an open source dataset by Chowdhury, et al.; Rahman et al. (2020). I start with a set of 96 radiomic features and reduce it down to a list of 10 definitive features. These radiomic features almost perfectly differentiate between radiographs with COVID-19, viral pneumonia, and no pathology, with an accuracy of 97.0%. This reduced model can help hospitals with overcrowding caused by COVID-19 by using these specific features in radiomic analysis to determine which patients have COVID-19 and are at the highest risk. At the moment, detection methods of COVID-19 such as PCR tests, and even the rapid tests take a significant amount of time for the results to be returned. In fact, PCR tests take 24 hours at the minimum, with rapid, take-home tests just slightly shorter. Additionally, sending chest radiographs to radiologists for analysis takes time, and also places excess stress on the radiologists. In comparison, radiomics takes a matter of seconds and is a completely automated process. Current technology relies on time-consuming methods such as the bacteria growth in conventional COVID-19 tests, and the human examination that radiologists carry out; in this study, radiomics is examined as a functional alternative.
Presented by
Arohi
Institution
Okemos High School
Keywords
Radiomics, Efficiency, Hospitals, Congestion, Rapid detection, Resource conservation, Radiographs, COVID-19, Analysis
Effect of bacterial diet on reproductive aging in C. elegans
Alex
Abstract
Women are born with all of their eggs, and as the eggs age, infertility and birth defects increase. One factor that may have a role in protecting eggs as they age is the formation of protein granules. C. elegans is a useful model system to study the role of protein granules in eggs. In the eggs of old-aged C. elegans, referred to as worms, the MEX-3 protein forms granules that may maintain the quality of the eggs. In addition, when the worms are fed certain strains of E. coli such as HB101 or HT115, they reproductively age faster than when they eat the standard OP50 bacteria. I hypothesized if worms are fed HB101 or HT115, protein granules will form earlier in eggs than in worms fed OP50 bacteria. The first experiment compared HB101-fed worms to OP50-fed worms. The second experiment compared HT115-fed worms to OP50-fed worms. I used worms in which the Green Fluorescent Protein was fused to the MEX-3 protein, and this allowed me to view whether MEX-3 protein granules formed in eggs of living worms using a fluorescent microscope. The worms were imaged when worms were 2-, 3-, and 4 day-old adults, and three trials were done for each experiment. Images were captured using a fluorescent microscope, and eggs in each worm were scored as having formed protein granules or if the protein remained diffuse throughout the cytoplasm. Statistically significant differences were seen: 4-fold more HB101-fed worms had granules on day 4 than OP50-fed worms, and 5-fold more HT115-fed worms had granules on day 3 than OP50-fed worms. The data support my hypothesis and show that different diets can be sensed, and adapted to, at a cellular level. Since some bacteria cause faster reproductive aging, the eggs may need to form granules earlier to protect the quality of the eggs.
Presented by
Alex
Institution
Mount Pleasant High School
Keywords
Aging, reproduction, C. elegans, diet
Designing and Constructing a 3D Printer for the Additive Manufacturing of Soft Matter Hydrogels
Henry Wolf
Abstract
3D printing, a method of additive manufacturing, involves laying down thin paths of material using a moving extruder in order to slowly build up a three-dimensional form layer-by-layer. While 3D printers typically use thermoplastics to build rigid parts, the technique can be applied to a variety of different materials with niche uses. Hydrogel 3D printers are an emergent medical technology that could eventually facilitate the manufacturing of replacement human tissue structures; a 3D printer could recreate the intricate internal structure of a realistic organ where traditional gel molds cannot. The goal of this engineering project is to design, build, and test a 3D printer specifically designed to form models out of 88% hydrolyzed polyvinyl alcohol gel. The printer features several unique components, including a custom syringe pump extruder for dispensing measured quantities of gel and a Peltier-stage-cooled printing surface for initial layer adhesion. The device was capable of printing single-layer geometries out of gels of varying concentrations. However, additional layers were largely unsatisfactory due to the time required to freeze the gel solid. The printer was a successful proof-of-concept for the process of printing hydrogels using a cryogenic build plate, but requires still more adjustment.
Presented by
Henry
Institution
Saginaw Arts and Sciences Academy
Keywords
3D Printing, Additive Manufacturing, Hydrogels
An Evaluation of the Effects of d-α-Tocopherol and Amygdalin on Regeneration of Dugesia tigrina
Armaan Mahajan
Abstract
Cancer is not an unknown killer. This disease is the second leading cause of death in the world. Cancer is the uncontrolled replication of cells within the body and does not respond to signals that regulate the growth of most cells. Many cancer cells have a defect in the p53 gene, causing cells to lose the information needed to respond to signals to control growth. Although there are many treatments for cancer, there is no cure as of now. The most common form of treatment for various types of cancer is chemotherapy. Chemotherapy drugs prevent the division of cancer cells, effectively stopping the growth of cancer throughout the body. Along with these side effects, chemotherapy drugs can also target healthy cells in the body. Although chemotherapy is the most common form of cancer treatment, there are still many harmful side effects of this treatment. It can result in lung tissue damage, heart problems, infertility, and nerve damage. Because of these side effects, many people have switched to dietary supplements to help prevent cancer. Some of the most discussed dietary supplement cancer treatments are d-α-tocopherol (vitamin E) and amygdalin. The objective of this study is to compare the rates of regeneration of planarian (specifically Dugesia tigrina) cells when exposed to vitamin E and amygdalin. Planarians were divided into two pieces using a razor and then exposed to various dosages of vitamin E and amygdalin. The rate of regeneration was then viewed and measured under a simple light microscope. Vitamin E managed to only slow down the rate of regeneration, while the highest dosage of amygdalin managed to stop regeneration from occurring. This supported the hypothesis that stated that amygdalin would prevent regeneration more effectively than vitamin E.
Presented by
Armaan
Institution
Saginaw Arts and Sciences Academy
Keywords
Cancer, Regeneration, Animal Sciences
Powering a Toothbrush without a Battery
Niloy Islam and Mohammad Islam
Abstract
In this project, a toothbrush is powered by using the Seebeck Effect through Peltier modules and hot water. This idea replaces the use of the traditional batteries and the charging system. Also, batteries contain heavy metals which could be detrimental to health and the environment. The battery charging system is not universally usable which makes this idea an attractive alternative. Peltier modules have a hot and cold side and produce voltage by a temperature difference between the two sides, known as the Seebeck Effect. Peltier modules were attached around a container holding hot water with the hot face inside. Heat sinks were used to spread the heat outside into open air to maintain a temperature difference for a given period of time. Each Peltier module is a voltage source that needs to be connected properly to obtain sufficient current and voltage, thus powering the brush for at least 120 seconds. The voltage (V) and current (A) were measured by using two digital multimeters. Shortly after hot water was poured into the Peltier reservoir, data was collected by creating a video of the toothbrush running. Ice on the outside and boiling water inside made the longest running time for the brush. More Peltier modules in series and more parallel paths also helped in operating the toothbrush longer. Tap water and normal ambient air made the brush run. Therefore, the Peltier modules and hot water worked effectively in powering a toothbrush without a battery.
Presented by
Niloy
Institution
Saginaw Arts and Sciences Academy
Keywords
Hot water, Peltier modules, Toothbrush
▼ 6th Award Back to top
Smart Vaccine Transportation and Storage Device
Charlie Nielsen, Catherine Wang, Uma Kale, Diane White
Abstract
With a raging pandemic, and existing deadly diseases, it is crucial that people are vaccinated. The active ingredients in a vaccine which help with immunity, degrade over time. There is a time and temperature relationship which the degradation follows and if the vaccine degrades too rapidly it is not able to be used for as long, and may not be effective when administered. It is vital that vaccines are closely monitored and stored properly when being transported so that they have the intended efficacy. In this project, Oral Poliovirus Vaccine was used as a model. It is one of the most temperature sensitive vaccines and a suitable representative for other sensitive vaccines. A microprocessor based monitoring system was developed using the known degradation kinetics of this vaccine to predict remaining usable shelf life and potency based on the environmental temperature of the vaccine. The device also monitors and calculates real time mean kinetic temperature, which is a way to assess how controlled the last mile cold chain vaccine transportation and delivery is. These calculation techniques can be applied to any vaccine or drug such as COVID 19 vaccine for which degradation kinetics are known. The usage of this device can help companies save money by optimizing vaccine storage and transportation conditions. More importantly, it prevents inactive vaccines from being utilized on patients.
Presented by
Evelyn & Anjali
Institution
Herbert Henry Dow High School
Keywords
Vaccines, Kinetic Degradation, Potency, Arrhenius Equation, Cold Chain, Mean Kinetic Temperature, Arduino
Analyzing the Effectiveness of Benford’s Law at Detecting Election Fraud using 2020 Election Data
Leland
Abstract
The 2020 election cycle was one of the most controversial elections in the history of the United States. After Joe Biden was declared president-elect on November 7th, 2020, attacks on election integrity skyrocketed from those of whom supported the Republican candidate. One claim that opponents of election results made was that vote counts in certain counties did not match a mathematical principle known as Benford’s law. Benford’s law is named after American electrician and mathematician Frank Benford (Canadian engineer Simon Newcomb discovered the concept, Benford rediscovered and legitimized it). Benford’s law states that in a large data set, the likelihood that any of the numbers start with the number one is much higher than any of the other numbers (2-9). Benford’s law can be applied to many different data sets, such as electricity bills, stock prices, house prices, and other sets of data, including election data.
The goal of this experiment was to determine whether or not Benford’s law could detect election fraud - and, if so, were there discontinuities in the 2020 election data. First, data was analyzed from people asking similar questions, to see if their results were conclusive in any way. Observing other projects similar to this one also gave ideas of how to analyze the data fairly and properly. Afterwards, a subject of analysis was chosen. For this project, 2020 election data from all 159 counties in Georgia were used. For each county, the vote totals for the two primary candidates were written down and tallied based on the first digit of each number. The tally results and percentages were compiled into line and bar graphs, respectively. They were graphed against the expected outcome. The hypothesis was not supported, as the data does not match the expected outcomes for Benford’s law to a reasonable degree.
Presented by
Leland
Institution
Saginaw Arts and Sciences Academy
Keywords
mathematics, statistical analysis, elections
Effects of Essential Oils on the Growth of Staphylococcus Aureus
Abstract
Antibiotic-resistant superbugs have served as a major worldwide issue for many years due to their strength. One of the most common ones, Staphylococcus aureus, is a leading cause of various bodily pathologies, and has become harder to treat over the years. Essential oils have proven to be significantly effective antimicrobial agents against many bacteria. The purpose of this experiment was to determine which essential oil: cinnamon, clove, tea tree, or thyme, would be the most effective in preventing the growth of S. aureus. Vancomycin is an antibiotic that was used as a comparison. Three different concentrations of each substance were used: 5 µl/disk, 10 µl/disk, 20 µl/disk. The hypothesis was that cinnamon oil would be the most effective, whereas tea tree oil would be the least effective. 100 µl of S. aureus were inoculated across a Mueller Hinton agar plate. Then, impregnated discs of cinnamon oil, clove, tea tree, thyme, and vancomycin, were placed on the agar plate, along with distilled H2O as a control. The plates were placed in an incubator at 37℃ for 24 hours, and the diameter of the zone of inhibition (ZOI) around each disk was measured to nearest millimeter. Three trials were conducted for each substance, and the mean was calculated for each set of trials. The results were that cinnamon oil was the most effective, with a mean ZOI of 17.3 mm, and clove oil was the least effective with mean ZOI of 9 mm, both at 20 µl/disk.
Presented by
Harman
Institution
Saginaw Arts and Sciences Academy
Keywords
Antibiotic resistance, bacteria, essential oils
Downregulation of EMP2 and PID1 cell proliferation genes in Lung Cancer
Aanchal
Abstract
Lung cancer is one of the leading causes of cancer deaths in the United States. Although there are many treatments, they can be expensive and have long-lasting effects. Being able to treat cancer-based on a Personalized Medicine approach could revolutionize medicine. The main purpose of this study was to identify genes downregulated in Lung cancer patients. Through using the Gene Expression Omnibus (GEO), the publicly available dataset GSE10072 was found to be related to Lung cancer. The dataset compares healthy control patients (with healthy lungs) to Lung cancer patients (adenocarcinoma in the lung). Using GEO2R, many genes were found to be downregulated in the Lung cancer group. To further analyze the differentially expressed genes (p-value <0.05), String-DB was used to find commonly grouped genes for many biological processes. The biological process of the Regulation of Cell Population Proliferation had 43 differentially expressed genes associated with it. It was found that the two genes out of the 43 were of interest - EMP2 and PID1 - both downregulated in the Lung cancer tumor group. Both EMP2 and PID1 share roles in cell proliferation, with PID1 specifically having functions in preadipocyte proliferation. These genes hold importance in tumor growth in Lung cancer since they are related to the Regulation of Cell Population Proliferation pathway. Since EMP2 and PID1 were downregulated in the tumor group, they are sure to have some role in the regulation of tumor growth. Identifying these genes now can provide insight into the future of Lung cancer and Personalized Medicine.
Presented by
Aanchal
Institution
miRcore
Keywords
Personalized Medicine, Genomics, Lung cancer
▼ 7th Award Back to top
Binding of Curcumin and β-Lactoglobulin and the Effect of Curcumin on Fibrils Formed by β-Lactoglobulin
Authors: Sharmitha Bandla, Mentor: Matthew Miller
Abstract
The goal of this project is to see if Beta-Lactoglobulin (β-Lg) and curcumin can bind and if curcumin can reduce the amount of fibrils formed by β-Lg. β-Lg is a major whey milk protein found in dairy products, such as milk, cheese, yogurt, etc, that causes allergies in children. Curcumin is a chemical that has many benefits for health and has antioxidant effects as well. The hypothesis was β-Lg and curcumin would bind, and that
β-Lg fibrils will be reduced in the presence of curcumin. A spectrofluorometer was used to measure the fluorescence intensity. The wavelength used for testing was 340 nm. If fluorescence levels are higher after adding curcumin to Beta-Lactoglobulin, then binding has occurred because when they bind, the polarity of their environment changes resulting in a higher fluorescence emission intensity. The fluorescence level of
β-Lactoglobulin (with no curcumin) is 90. The average fluorescence levels after adding curcumin were 360, 355 and 330. To determine if curcumin reduced the amount of fibrils, a spectrophotometer was used and the wavelength was also set at 340 nm. After the addition of curcumin, the absorbance values went down, and this corresponds to a decrease in the number of fibrils of β-Lg in either size or number. Based on the data collected, the hypothesis was supported and we can conclude that β-Lactoglobulin binds with curcumin and that curcumin reduces the amount of fibrils formed by β-Lg. This project may have beneficial applications for the food industry and implications regarding neurodegenerative disorders.
Presented by
Sharmitha
Institution
Saginaw Arts and Sciences Academy
Keywords
Medicine, health, spectrofluorometer, spectrophotometer, curcmin, β-Lactoglobulin
How Does Temperature Affect the Iodine Clock Reaction?
Author: Caleb; Mentors: Rob and Michelle
Abstract
Chemical kinetics is a fascinating field of chemistry, but it is a very hard field to study. This is due to the fact that most experiments and reactions occur so quickly that it is impossible to measure them unless you have specialized chemistry equipment. This is not the case with the iodine clock reaction. The reason the clock reaction is useful is because it is a reaction that is easily measurable with household equipment, like a stopwatch. It allows people to easily see the effects of chemical kinetics, without specialized equipment. It compares the reaction rates of varied temperatures; showing how a solution behaves differently with altered conditions.
Presented by
Caleb
Institution
Homeschool
Keywords
Chemistry, Reaction Rates, Iodine Clock Reaction, Chemical Kinetics
Comparing Methods of Recoil Measurement
Author: Addison Wagner Mentor: Dr. David Allan
Abstract
All guns have recoil energy after they shoot. Recoil pads redirect this energy to relieve pressure from the shoulder. Different materials cushion these pads, and different methods can test these materials. In this experiment, recoil pad materials (air, gel, hydraulic fluid, Sorbothane) were tested for their effectiveness, and testing methods (pendulum, force meter and gun, company tests) were compared for their accuracy. The hypotheses were: the hydraulic pad would be the most effective, and the pendulum would be accurate compared to results from a force meter and the pads’ companies. To test these, a pendulum with 3.2 kg and 25.0 kg bobs was built with 1.2 m beams. The pads were attached to the 3.2 kg bob and dropped from a 90-degree angle to simulate a gun hitting a shoulder, each trial being filmed. Using centimeter graph paper behind the pendulum with the distance formula, the distance the bob was pushed back with each pad was found. However, the pendulum was not accurate compared to the other tests and the results could not be directly interpreted. Instead, effectiveness was inferred from the idea that the farther the pendulum was pushed, the more cushioning the pad is. Therefore, the hydraulic pad was the most effective, pushing an average of 61.9 cm. Following was the air pad (61.4 cm), gel pad (52.5 cm), and Sorbothane pad (50.2 cm). These results did not support the hypothesis that the pendulum would be accurate, however, they did support the hydraulic pad being the most effective.
Presented by
Addison
Institution
Saginaw Arts and Sciences Academy
Keywords
Recoil, Materials Science
How does tire pressure and weight affect rolling resistance?
Caitlyn Dickerson
Abstract
In this experiment the question was answered, how do tire pressure and weight affect rolling resistance? The goal of this experiment was to prove that tires with lower psi and more weight will have more rolling resistance and tires with higher psi and less weight will have less rolling resistance. To prove this hypothesis a bike, three volunteers, a spring scale, and a pressure gauge/bike pump were used. The spring scale was used to pull the bike with the volunteer on it. The results of this project mainly proved that more weight and lower tire pressure have more rolling resistance than less weight and higher tire pressure.
Presented by
Caitlyn
Institution
Calvary Baptist Academy, Midland MI
Keywords
psi, weight, bike, Newton, spring scale
▼ Honorable Mention / Research Fellow Back to top
Effects of Vitamin C and Selenium on Invertase
Helsa
Abstract
The objective of this experiment was to determine the effects of Vitamin C and Selenium on Invertase. The information obtained from this experiment can aid the general public in making healthier choices. Before experimentation, it was hypothesized that Vitamin C would have a greater effect on Invertase as opposed to Selenium. To test this hypothesis, a spectrophotometer was used to test the absorbance of these three test subjects, as well as distilled water. Invertase was extracted from yeast, and Vitamin C and Selenium tablets were crushed and mixed into 40 milliliters of distilled water. Absorbance was measured at 220, 240, and 260 nanometers. After data was collected, concentrations of Selenium and Vitamin C were found in order to determine results. A trendline was created based on the absorbance values of Invertase and Distilled water. Average concentrations of Selenium and Vitamin C were found using the trendline and concentrations were graphed. It was concluded that the hypothesis was supported, and that Vitamin C had a greater effect on Invertase than Selenium.
Presented by
Helsa
Institution
Saginaw Arts and Sciences Academy
Keywords
Vitamin C, Selenium, Invertase, Absorbance, Medicine, Health, Biomedical
Using the N.E.A.T. Genetic Algorithm as a Substitute for a Supervised Learning Algorithm
Rayen
Abstract
The artificial neural network is a subset of machine learning loosely inspired by the biological neural networks that constitute human brains. They enable one to train a computer to make predictions or decisions without being explicitly programmed to do so. They have numerous applications, from predicting data to identifying items from images to even controlling self-driving cars. There is an abundance of different algorithms through which they can be optimized for their intended task. A fairly recent method of optimization is the Neuro Evolution of Augmenting Topologies or N.E.A.T. genetic algorithm. While the N.E.A.T. algorithm is considered a reinforcement learning technique, this experiment seeks to prove that it can be used as a substitute for backpropagation, a supervised learning technique. The experimental procedure entails: using a program capable of training neural networks through the backpropagation algorithm to train a neural network with an arbitrary data set for ten trials, recording the accuracy of the model for each trial, finding the average accuracy as well as the standard deviation, repeating the previous steps with a program capable of training neural networks through the N.E.A.T. algorithm, then comparing the average accuracies with their standard deviation in mind. The N.E.A.T. algorithm had an average accuracy of 98.80% with a standard deviation within a reasonable amount, while the backpropagation algorithm had an average accuracy of 99.97% with a standard deviation within a reasonable amount. The results of the experiment showed that the N.E.A.T. algorithm had an average accuracy of just 1.17% less than backpropagation, proving that it could be used as a viable substitute for backpropagation, and therefore proving the hypothesis to be supported.
Presented by
Rayen Aouadi
Institution
Saginaw Arts and Sciences Academy
Keywords
Machine learning, deep learning, artificial intelligence, AI, A.I., neural networks, artificial neural networks, N.E.A.T., NEAT, genetic algorithm, backpropagation, gradient descent, supervised learning, reinforcement learning, unsupervised learning, computer science, programming, coding
Sweat It Out: Engaging Aerobic Exercise as a Biological Solution to Improved Mental Wellbeing
Christopher Wells
Abstract
https://docs.google.com/document/d/1t9xk6Ryls4iN_iu3c38Dut6yi8K-qaN0YIBXSAXYjGE/edit?usp=drivesdk
Presented by
Katrina
Institution
Wells Family
Keywords
Aerobic exercise, Mental Health, Lifestyle
Testing the Phytoremediation of Lead in Salix Species.
Vanessa
Abstract
This experiment is being conducted to further understand that some plants can phytoremediate, and in specific, phyto-accumulate toxins in contaminated soil, while thriving in the environment. Phytoremediation is a process in which contaminants are removed, immobilized, transferred or destroyed from soil or groundwater using different species of plants. These toxins can make it impossible to grow crops or allow for close human contact. Phytoremediation offers an environmentally friendly alternative to expensive soil clean-up. It was hypothesized that the Salix caprea Pendula used in this experiment will slowly phytoremediate, or more specifically, phyto-accumulate the lead in its soil. A lead solution was added to the soil of a Salix caprea Pendula plant, which was then tested to determine if the plant phyto-accumulated the heavy metal. The willow plant immediately performed phytoremediation on the lead to take it from the soil, showing a promising future for a cost-effective, environmentally friendly way to clean contaminated soil.
Presented by
Vanessa
<burkhardv@k12.spsd.net>
Institution
Saginaw Arts and Sciences Academy
Keywords
Environmental Science, Phytoremediation, Plants, Plant Science
Plant Growth and the Effects Of Caffeinated vs. Decaffeinated Coffee
Marie Hillebrand
Abstract
The question that this project was based on was, how does coffee, decaffeinated or caffeinated, affect the growth of sunflowers? The hypothesis that was thought of was “if a ½ cup of half and half caffeinated coffee and water solution is used to water sunflower seeds every week, then the growth of the sunflowers will increase the most compared to the growth of sunflowers that were watered normally and watered with decaffeinated coffee.” This project was conducted by dividing nine sunflower seeds into three different groups, each with three plants in them, and then watering each group with either water, a half and half solution of regular black coffee, or a half and half solution of decaf coffee every five days for four weeks. At the end of each week, the growth of each plant was measured by recording the height, width, leaf number, and coloration of each plant. Overall, the control group exhibited better growth compared to the caffeinated and decaffeinated group, especially when the height is considered. A reasonable conclusion based on this project is that watering chocolate sunflowers normally is better than giving them a half and half solution of coffee.
Presented by
Miranda
Institution
Calvary Baptist Academy
Keywords
Biology, Botany, Sunflowers, Coffee
Emission Recognition
Emma
Abstract
Diesel Exhaust Fluid (DEF) is used in industrial machines and trucks to transform nitrogen oxide emissions into water and nitrogen. DEF is composed of 32.5% urea and 67.5% deionized water.
The purpose of this experiment is to determine the ratio of urea in the DEF and the ratio of urea on the DEF product label. The hypothesis of this experiment was that Love’s DEF will have the lowest urea percentage average compared to the urea ratio on the DEF product label and Blue DEF will have the closest urea percentage average to the urea ratio on the DEF product label. Prior to the experimentation, the ratio of urea on the DEF product label will be recorded. In each trial ½ mL of each DEF brand will be tested by a refractometer to get a urea percentage. After all trials are completed, the average urea percentage of each brand of DEF will be calculated. Then the urea percentage averages of each brand of DEF will be compared to the ratio of urea on the DEF product label. Upon conclusion, it was determined that Love’s DEF, Motorcraft DEF, and Victory DEF had the closest ratio of urea compared to the ratio of urea indicated on the DEF product label. Blue DEF had the least closest ratio of urea compared to the ratio of urea indicated on the DEF product label. It is very important to keep industrial machines and trucks running properly to keep up with the demand for products.
Presented by
Emma
Institution
Saginaw Arts and Sciences Academy
Keywords
Chemistry, Diesel Exhaust Fluid, Urea Percentage
Testing the Ability of Machine Learning to Analyze Characteristics of Human Artwork
Joseph
Abstract
The goal of this project is to utilize machine learning to analyze aspects of paintings, and determine which characteristics of artwork can be easily assessed. Paintings were sorted into categories based on the time period, genre or artistic movement, an expected emotional response, and major themes. It is expected that that machine learning will easily be able to determine the time period a painting is from, and will also be able to determine the genre, but it will struggle with or be unable to interpret emotions and themes in the paintings. Some similar studies have been conducted, but not ones that measure the same characteristics of paintings as this one. Tensorflow was used to test different machine learning models. Once one was decided upon, it was tested with all four sorted datasets. The expected outcome did not occur; emotional association was actually the easiest for the computer to determine, and the algorithm struggled greatly with predicting the time period.
Presented by
Joseph
Institution
Saginaw Arts and Sciences Academy
Keywords
machine learning, computer art, painting, computer science
Electro-Adhesion Water Filtration
Dev
Abstract
Abstract
Electro adhesion can remove heavy metals Chromium 6, Lead, Mercury, Arsenic from water. It can also remove microplastic. Microplastics and heavy metal in water can cause several severe diseases like brain cancer, lung cancer, anemia, heart and kidney problems. Electro-Adhesion utilizes the difference in charges that exist between a surface and a particle in aqueous solution, where a charge is built up by the double layer effect. Zeta potential is the measurement of the driving force between the particle and the fixed surface, it can attract or repel the two. Most bacteria and most other particles are electronegative in water. Smaller particles also tend to become more electronegative. So an electropositive fixed surface would be far more effective at attracting and retaining particles than one that is electronegative. Another factor is the area of solid surface that is exposed to the particles in the fluid. One with a large surface area can support more electropositive charges and therefore adsorb more particles. In this experiment a filter that uses electro-adhesion is used and a device is created to put the filter in and different samples of water from different ponds and faucets. The hypothesis was that the filter can remove waste particles up to 0.02 microns as well as >99.997% at 0.2 microns and >99.997% microns (which is virus level) in a pH range of 4 to 9. The results were that my hypothesis was supported and the data were collected by counting the number of particles before and after the experiment. One major finding was that a sample of pond water with 600 particles was found before filtering and after filtering, the highest number of particles were 3. This shows that the filter was able to clean it, this was the worst sample of water out of all 6 samples and the color changed from black to clear. So electro-adhesion can filter out water, these filters can filter water to make it clean enough for human use.
Electro adhesion can remove heavy metals Chromium 6, Lead, Mercury, Arsenic from water. It can also remove microplastic. Microplastics and heavy metal in water can cause several severe diseases like brain cancer, lung cancer, anemia, heart and kidney problems. Electro-Adhesion utilizes the difference in charges that exist between a surface and a particle in aqueous solution, where a charge is built up by the double layer effect. Zeta potential is the measurement of the driving force between the particle and the fixed surface, it can attract or repel the two. Most bacteria and most other particles are electronegative in water. Smaller particles also tend to become more electronegative. So an electropositive fixed surface would be far more effective at attracting and retaining particles than one that is electronegative. Another factor is the area of solid surface that is exposed to the particles in the fluid. One with a large surface area can support more electropositive charges and therefore adsorb more particles. In this experiment a filter that uses electro-adhesion is used and a device is created to put the filter in and different samples of water from different ponds and faucets. The hypothesis was that the filter can remove waste particles up to 0.02 microns as well as >99.997% at 0.2 microns and >99.997% microns (which is virus level) in a pH range of 4 to 9. The results were that my hypothesis was supported and the data were collected by counting the number of particles before and after the experiment. One major finding was that a sample of pond water with 600 particles was found before filtering and after filtering, the highest number of particles were 3. This shows that the filter was able to clean it, this was the worst sample of water out of all 6 samples and the color changed from black to clear. So electro-adhesion can filter out water, these filters can filter water to make it clean enough for human use.
Presented by
Dev
Institution
Saginaw Arts and Sciences Academy
Keywords
Water Filtration
The Antibacterial Effect of Commonly Consumed Beverages and Mouthwash on Streptococcus mutans
Anthony
Abstract
Intro: Dental caries are one of the most common and costly infections in the world. One of the major contributors to these caries is Streptococcus mutans that can alter the local environment thereby creating a favorable niche for other species to thrive. The human diet also plays an important role in dental caries; some substances might provide specific bacteria with some nutrients that would enhance their growth over others playing therefore a role in their selection and over multiplication.
Procedure: The antibacterial activity of the oral substances against Streptococcus mutans was determined using the macrodilution method and expressed as a percentage of substance in the solution. The nine substances that were tested are: coffee, tea, energy drinks, wine, vodka, lemon juice, orange juice, mouthwash and antiseptic solutions. For this, brain heart infusion broth and agar were used. Ampicillin was used as control.
Results: Of all the nine substances tested, wine, vodka, mouthwash and strepsils affected the growth of S. mutans at different concentrations. While the energy drink, orange juice, coffee and tea were not associated with detectable activities in the conditions of the experiments. Among those that exerted anti-Streptococcal activity lemon juice had the least inhibition (at 9.38%) followed by Strepsilis (5.21%), wine (4.17%), mouthwash 1 (1.43%), vodka (1.43%) and mouthwash 2 (0.85%).
Conclusion: The effect of different beverages and mouthwash on the oral microflora should be taken into consideration when in treatment of dental caries and oral infections.
Presented by
Anthony
Institution
Saginaw Arts and Sciences Academy
Keywords
Streptococcus mutans, antibacterial effect, MIC method, beverages, mouthwash.
Refraction of Light in Concentrations of Differing Molarity
Ben
Abstract
Presented by
Ben
Institution
Saginaw Arts and Sciences Academy
Keywords
DELETE
Akshanth
Abstract
Presented by
Akshanth
Institution
Saginaw Arts and Sciences Academy
Keywords
Cocrystal
Effects of Initial Mycelium Injection on Oyster Mushroom Growth
John Barnes (mentor)
Abstract
This project is an exploration into mycelium growth. The purpose of this project is to look at to determine the quantified impact that the original amount of injected mycelium has on a mushroom culture. It was hypothesized that the mass of mushroom growth (in grams) would increase with the initial amount of injected mycelium (in cm²). The hypothesis was tested by mixing a substrate, injecting different amounts of mushroom culture into the jars, and then leaving those to grow in a warm and dark terrarium. The hypothesis was not supported by the data, as the resulting mushroom growth differed so widely that trends were not visible in the data.
Presented by
Julia
Institution
Saginaw Arts and Sciences Academy
Keywords
Mushrooms, Mycelium
Examining the Law of Large Numbers as It Applies to Various Amounts of Possible Outcomes
Nathaniel Watson
Abstract
The objective of the experiment was to determine if the amount of time it takes for the law of large numbers to come into effect is affected by the amount of possible outcomes. To test this, a digital dice-rolling simulation was used to generate 1000 random outcomes per number of possible outcomes ranging from two to six. The means of each set of outcomes was recorded for every 100 outcomes generated, and the resulting data was graphed and used to come to a conclusion. It was hypothesized that the law of large numbers would take longer to come into effect for the simulations with fewer possible outcomes. This hypothesis was not supported by the data, and it was concluded that the law of large numbers takes less time to come into effect for simulations with fewer possible outcomes.
Presented by
Nathaniel
Institution
Saginaw Arts and Sciences Academy
Keywords