Identifying an Effective Semantic Segmentation Algorithm for Deforestation Analysis

The purpose of this project is to implement a method for analyzing changes in land cover due to deforestation in the United States. Previously, the USDA used decision-tree classification in semantic segmentation (i.e., labelling of each pixel in an image with a category label) to assess land cover change. However, deep learning has recently become popular, and it has yet to be applied successfully to deforestation analysis. The proposed algorithm incorporates convolutional neural networks, but with an additional filter, called atrous convolutions. These filters are dilated in order to achieve a broader view of the scene or image.

How Changes in Gene Expression Affect Phenotypes and Clinical Outcomes for Head and Neck Squamous Cell Carcinoma

Cancer is a disease caused by the uncontrolled division of cells in a part of the body. In general, cancers arise as a result of DNA changes in the genomes of cancer cells. Genes encode proteins, and proteins then dictate cell function. The genes that are “expressed,” or turned on, in a particular cell determine what that cell can do. As a result, one possible cause of head and neck cancer, a subtype of cancer, is differences in gene expression levels between patients with cancer and patients without cancer. This research project will analyze gene expression data from head and neck cancer patients using the Seurat package in the programming language R to identify marker genes that can later be suppressed with gene editing technology.

How a Systematic Process of Collecting and Presenting Evidence in Court Can Be Installed in the United States to Increase the Reliability of Forensic Science

“Forensic Science is the application of scientific principles and techniques to matters of criminal justice especially as relating to the collection, examination, and analysis of physical evidence” (Webster, 2018). Across the United States, forensic science plays a critical role in how criminal cases are both investigated and decided. One major issue is that there have been flaws in how evidence is collected and presented, causing some cases to have skewed evidence leading to false imprisonment. The problem I intend to tackle is how we can install a baseline protocol for how to collect and present evidence along with how we can filter out evidence that was poorly collected.

How 4D Printing, Smart Materials, and Self-Assembly Can Be Implemented Together to Protect Humans in Hostile Situations While in Space

The aim of this project is to figure out if there is any overlap between three fields of nanotechnology: 4D printing, self-assembly, and smart materials. 4D printing is the process of 3D printing objects that change shape as the conditions of the object's environment are changed. Self-assembly is the process of engineering conditions so that certain organic molecules come together to form ordered macromolecules. Smart materials are those that have different properties in different conditions. Furthermore, the project aims to find any applications of these three technologies that can be used to make space travel more efficient and safe.

An Empirical Approach to the Measurement Problem of Quantum Mechanics

I am exploring measurement collapse, where a particle tends to instantaneously fall into a definite state only after being observed by a measuring device or conscious being. Since the 1920s, many interpretations on how to explain how and why this behavior occurs have been presented. However, the quantum mechanics community has not been able to determine which interpretation is most accurate. This project will focus on one of these interpretations, "simultaneous localization," which holds that collapse occurs randomly and without the need for measurement. This project will utilize the IBM Q Experience 5-qubit quantum computer to empirically simulate the simultaneous localization measurement process. To simulate this interpretation, we will observe what occurs to the uncertainty in spin of two multi-particle entangled systems with opposing spin. In this particular situation, the simultaneous localization interpretation and the accepted orthodox interpretation of measurement differ. If we can analyze which interpretation more closely matches particle behavior in this scenario, we can get begin to determine which interpretations are most accurate.

Applications of AI in the Medical Field

As artificial intelligence is used more and more, people are starting to explore how AI can be used on the medical field. However, AI still makes errors, which means that we must improve AI and tailor its application so that it can be as effective as possible. If used effectively, this can help improve the accuracy and efficiency of diagnosis and other processes, drastically improving medical research.

Analyzing the optogenetically stimulated neurons in the Bed Nucleus of the Stria Terminalis (BNST) region of mice through behavioral testing and cell analysis

This research aims to simulate the stimulation of an areas of the brain so that it can be used as a template for other experiments. Building upon previous studies that focused on how the reward centers of the brain are affected by stimuli from other regions of the brain, the study research will involve analyzing the optogenetically stimulated neurons in the BNST (Bed Nucleus of the Stria Terminalis) region of mice through behavioral testing and cell analysis. The behavioral testing will be used to analyze whether or not the stimulation was rewarding.

How can a decision tree help guide patients through consideration for the different walking devices available to them?

The goal this project is to create a decision tree that helps patients see the different factors to consider for the different walking devices available to them. The end product will hopefully give patients a better idea of the pros and cons of each walking device, such as cost, location, compatibility, etc.

What are the pros and cons of stem cells and gene therapy when used to cure blindness?

This research project aims to provide a platform for people who are interested in stem cells to cure blindness. The platform will include different treatments, success stories, and risks to the operations. The studies will be pertaining to common eye diseases such as glaucoma. The platform will be unbiased and will only provide information.

What strategies can Palo Alto Unified School District (PAUSD) use to effectively reduce the amount of waste created and increase the efficiency in which waste is sorted (into recycling, compost, and landfill bins) on elementary, middle school, and high sc

This research project’s ultimate goal is to create a formal plan containing suggestions as to how the PAUSD can reduce the amount of energy consumption, both natural gas and electricity, at PALY. In order to do so, the project will include four types of data collection: surveys, interviews, energy monitoring at PALY, and analysis of external and retrieved data.

Pages

STEAM

Subscribe to RSS - STEAM