STEAM

0

AI Algorithm that Calculates the Heart Rate from an Electrocardiogram (ECG)

My mentor will provide data sets of electrocardiograms that his company has collected. Using knowledge of machine learning that I will gain through background research, I will create an AI algorithm that will "learn" the relationship between an electrocardiogram and the heart rate. Essentially, I will feed my algorithm sets of electrocardiograms with corresponding heart rates, and the algorithm will find a relationship between them. When fully trained, my algorithm will provide an accurate heart rate for any electrocardiogram that I give it.

0

How to Optimize and Naturalize Computer-Designed Images

Utilizing machine learning, computers can now be taught to generate real images. This project uses a general adversarial network (GAN) to generate images. A GAN is composed of two neural networks: one plays the painter and one the judge. The painter uses responses from the judge to improve its painting, eventually generating life-like images. However, getting a GAN to work as intended in this regard is notoriously difficult; one goal of the project is to improve the training process.

0

Testing Graph Theory's Usefulness in Creating New Materials with Expected Properties through Computational Experiments Through Quantum Espresso Computer Simulations

This project will explore whether graph theory can be used to control chirality and model the genesis of chirality, a molecule not superimposable on its mirror image. This will be determined by observing isomorphism, or whether they are corresponding or similar in form, in CoCalc. Also, the project will explore whether properties of nanosystems calculated in Quantum Espresso (i.e., magnetization, thermodynamics, and band structures) can be predicted through graph theory. Finally, there will be an investigation on whether graph theory can mathematically describe the growth mechanisms in graphene.

0

Modelling Control Factors For Mosquitoes In Many Environments

We will be creating a computer simulation that models how populations of mosquitoes are affected by certain vector control methods such as Wolbachia. This simulation will model a variety of methods and their effectiveness in various conditions and environments. The simulation will be interactive, with the ability to add factors in the process.

0

Effective Methods for a Counselor-Based School Mental Health Program to Prevent Additional Suicide deaths After a Public Suicide in PAUSD

Despite numerous attempts at preventing suicide, high-achievement communities are still struggling to design programs that are effective at preventing these deaths. This problem has negatively impacted both individuals and communities at large, since people are still suffering with mental health issues and don’t know where to seek affordable counseling and mental health services. Perhaps exploring past suicide data, and the mental health programs in those impacted communities, can point the way to a possible remedy for this problem.

0

Graphene Doping

We will examine how properties of graphene change as we dope them with pairs of metals. We are using Quantum Espresso to simulate how the particles interact with each other. We will also compare these changes in properties to other experiments to see how this can be used as a semiconductor or for other purposes.

0

Preventing Pedestrian-Involved Automobile Accidents Using Inexpensive Portable Warning Systems (PWS) Leveraging Technology Advancements in Sensors and Machine Learning Software

Automobile accidents are an ongoing epidemic in the United States. Pedestrian death rates and total automobile-related death rates have increased in recent years. Autonomous cars are the most anticipated solution to this problem; however, it will take a long time until all vehicles on the street are autonomous. I propose a secondary solution: to build a portable warning system (PWS) for pedestrians to reduce car crashes. My research project is to test whether it is feasible to build such a device using today's latest technology.

0

Harnessing the Power of the Ocean: Designing a Portable and Efficient Wave Energy Converter Prototype

I will design and build a prototype of a device that generates electricity using the power of ocean waves. It will be efficient and portable so that it can be easily be used by people who have been affected by natural disasters or do not have access to electricity. My prototype will involve a coil and a magnet that induces a current as it shakes.

0

Testing for a Link between Bacteriophages/Gut Bacteria and Parkinson's Disease

The recent discovery that bacteriophages may influence the development of Parkinson's disease sparked our interest in a less explored area. We investigated the relationship between bacteriophages and Parkinson's disease by interviewing researchers and professors, analyzing literature, and conducting an experiment studying the abundance of specific bacteriophages in varied environments. We hope that our research may inspire the growing science community.

0

Molecular Mechanisms Underlying Neurodegenerative Diseases Associated with Mutant Tau

Alzheimer's disease is the 6th leading cause of death in the United States, but currently, there is no cure (Alzheimer’s Association). To treat Alzheimer's disease, we must first fully understand the molecular mechanisms behind it and the pathological roles of the tau protein that classify diseases such as Alzheimer's disease as tauopathies. The tau protein has the normal role of attaching to and organizing microtubules, which are crucial for maintaining cell structure, transporting materials within cells, and separating chromosomes during cell division. Among other causes, mutations in the tau gene have been linked with pathological tau and the degeneration of neurons. I intend to research the effects of such mutations in a particular region of the gene, known as exon 10, on tau function and cell viability with both database and lab observational techniques.