FemPharma


Inspiration

This project comes from something deeply personal. Living with Crohn’s disease, I’ve faced countless hospital visits and medications. During one of my worst flares, I was overdosed — not because of negligence, but because the dosage guidelines were based on studies done mostly in men.

That moment made me realize something heartbreaking: women like me are invisible in so much of medical research. I didn’t want anyone else to go through that same fear, so I built FemPharma.


What it does

FemPharma is an AI-powered tool that detects gender bias in drug data and highlights risks where women may react differently than men.

  • Ingests FDA data, clinical notes, or manual inputs
  • Detects when dosages, side effects, or outcomes are skewed toward one gender
  • Provides visual, interactive reports that make treatments safer, fairer, and more personalized

How we built it

  1. Designed a pipeline to clean messy FDA and clinical drug data
  2. Used the TabFBN AI module to analyze structured tabular data and reveal hidden gender-based differences
  3. Built an interactive dashboard to visualize risks, compare male vs. female outcomes, and simulate different scenarios
  4. Packaged everything into a deployable app — so it’s not just code, but something real that patients, doctors, and researchers can use

Challenges we ran into

  • Health data limitations: real-world datasets are incomplete, male-skewed, and messy
  • Emotional weight: reliving my own hospital experiences while building the project was painful, but it reminded me why this matters so much

Accomplishments that we're proud of

  • Turning a deeply personal story into a solution that could impact millions
  • Building a prototype that not only detects bias but explains it visually and interactively
  • Proving that FemPharma represents both science and humanity

What we learned

  • Bias isn’t just a technical issue — it’s a human issue
  • It hides in the data but shows up in real lives, in hospitals, in the way women’s pain is dismissed
  • AI can be more than prediction — it can be advocacy and a voice for the overlooked

What's next for FemPharma

  • Expand datasets and integrate more FDA drug records
  • Make the tool usable by clinicians in real time
  • Create patient-friendly reports, so women can advocate for themselves
  • Work toward rewriting the medical story — one where women are no longer invisible in research, and care is truly equal

Share this project:

Updates