Hey there!

Hey, I'm Chris. This is where I'll ramble about topics that span medicine and computer science.

Many of my posts are work in progress and will be updated over time!

Latest Posts

Posts By Category

Similarity Search

Clinicians Who Code

My Work

  • Mere Medical

    meremedical.co

    A free, open source personal medical record app that makes it easier for patients to manage their own medical records in one place instead of across different online patient portals. Supports Epic, Cerner, and Allscripts by utilizing HL7 SMART on FHIR.

  • Exploring CQL Support in SDC Questionnaires

    github.com/cfu288/cql-lforms-proposal

    Exploring the feasibility of adding support for CQL in FHIR Questionnaires and to the NIH lforms npm package, enabling the execution of user-provided CQL expressions and libraries within questionnaires in the browser. Enabling this would allow for the creation of dynamic forms that can calculate the values of items in a form based on user-provided CQL.

  • What's In a Name: Evaluating Implicit Bias in OpenAI's GPT-3 Large Language Model in Medical Note Generation

    github.com/cfu288/gpt3-medical-bias

    An informal analysis of implicit bias in Large Language Model GPT-3.5 in the context of medical history note generation. Given only a patient name, does GPT-3.5 generate a different medical history note if provided a Caucasian sounding name vs and African-American sounding one? Early analysis revealed small but significant differences in past medical history and medication use in the generated notes.

  • XPC Hackathon Participant - AI in Personal Health Records

    hackathon.xprimarycare.com

    The prototype that resulted from a two-day hackathon hosted by X=Primary Care in the Patient Empowerment track to ideate and build out technical solutions to empower patients by helping them better understand their medical records using machine learning and AI.

  • Space Interns

    spaceinterns.org

    An online platform that connects underrepresented students interested in the aerospace field to various internship and job opportunities. Funded in part by the NASA New York Space Grant

  • Molecule Lipophilicity Regression and Prediction

    github.com/cfu288/ml-predict-lipohpillicity

    Utilized sklearn, RDKit, Pandas and more to perform regressions on a molecular dataset. Predicted lipophilicity values based on molecular structure with an r2 of 0.8 and RSME of 0.75 using a SVM.