• Projects

  • Co‑Designing Effective Human‑Algorithm Collaborations for Suicide Risk Assessment with Clinicians, Teens, and Families

    Spring 2021

    Early suicide risk detection is critical in suicide prevention, as the majority of suicide deaths between the ages of 10 and 24 occur from the first attempt. Given the challenges of assessing suicide risk based on patient self-reporting, the use of algorithmic decision-support systems (ADS) to support clinicians’ diagnostic capabilities has seen growing interest over the past several years. However, knowledge surrounding the effective and ethical implementation of modern ADS into clinical practice remains sparse. The purpose of this research was to inform the design of new ADS interfaces and training modules that support responsible and ethical use of suicide risk algorithms in clinical practice. Working with a team from CoALA Lab at CMU's Human-Computer Interaction Institute, I conducted a series of semi-structured interviews and participatory design activities with individual participants, in which they were asked to generate and evaluate new ideas for how these kinds of clinical decision support systems should be designed.
    Research Paper

  • Alexa MD

    February 2020

    Alexa MD uses a decision tree based structure and a voice assistant (Alexa) to streamline and automate the process of diagnosis and ease the burden on the healthcare industry. I focused on developing the back-end of symptom-disease relationships with Neo4j. We submitted Alexa MD at TreeHacks at Stanford University.
    Devpost

  • Foresight

    September 2019

    Designed to assist visually impaired individuals, Foresight is an armband that uses a haptic feedback system to guide a user's arm towards an object and ultimately enables them to pick it up. I focused on implementing the visual object detection with AWS Sagemaker. We developed ForeSight at PennAppsXX and placed 10th out of 251 teams.
    Devpost

  • Interactive Python Trainer

    November 2018

    My term project for 15-112 Fundamentals of Programming and Computer Science serves as an interactive study tool to help students learn how to properly construct functions in Python based on tasks and guided steps inputted by teachers. Because of the variety in prompts that teachers can provide, students also gain experience analyzing what individual lines of code do, identifying errors in code, and writing their own lines based on teacher prompts. Overall, this teacher-guided study tool helps to provide students with a more robust understanding of Python and greater confidence when constructing their own functions. This projects utilizes sqlite3 and nltk packages.
    GitHub

  • Investigating Correlations Between Environmental Factors and Bacteriophage Genomes

    July 2018

    As a participants of Pennsylvania Governor's School for the Sciences 2018, my team engaged in computational biology research related to the relationships between the environment and the genetic make-up of bacterial viruses. Specifically, I focused on clustering genomic data from the Actinobacteriophage Database using a k-means algorithm in Python, allowing us to identify trends within and between different clusters.
    Research Paper