MediBias

Inspiration

MediBias was inspired by a pattern we kept noticing: women often know something is wrong with their bodies long before the healthcare system acknowledges it. Too many symptoms are dismissed as stress, anxiety, or “normal,” only for serious conditions to be diagnosed years later. When we looked deeper, we realized this isn’t just a communication issue but really, it’s a data problem. Medical research has historically been done with male test subjects, leaving dangerous gaps in how female biology, hormones, and pain are understood. We wanted to build something that helps make those gaps seen and acknowledged, while also taking steps towards removing it.

What it does

MediBias is a data-driven tool that highlights systemic bias in medical research and clinical decision-making. It allows users to evaluate how inclusive a study, drug, or area of research is by looking at factors like gender representation, biological considerations, and whether sex-specific diagnosis or dosing was considered. Instead of diagnosing or assigning blame, MediBias helps surface what may be missing and encourages better questions, deeper listening, and more equitable care.

How we built it

We built MediBias as a web-based platform using HTML, CSS, and JavaScript, focusing on clarity and accessibility. The core feature is an Inclusivity Tracker that takes structured inputs and translates them into an easy-to-understand inclusivity grade. We designed the site to clearly explain the problem, provide evidence of bias, and show how data can be used to create accountability without replacing clinical judgment.

Challenges we ran into

One of our biggest challenges was scope. Systemic bias in healthcare is a massive issue, and we had to be intentional about focusing on what could be meaningfully built in a single day. Another challenge was responsibility, which is why we were careful to avoid creating anything that could be interpreted as medical advice or diagnosis, hence why we created this as a research tool instead. Creating the right balance between impact and ethical boundaries required a lot of thoughtful discussion.

Accomplishments that we're proud of

We’re proud that MediBias tackles a complex, sensitive issue in a way that is both accessible and responsible. In a short amount of time, we created a working tool with a clear narrative, a strong overall framework, and a design that makes systemic bias easier to understand. We’re especially proud of how the project turns silence and missing data into something measurable and actionable.

What we learned

Through building MediBias, we learned that bias in healthcare often begins long before the patient experience, but it starts in research design and data collection. On a technical level, we gained experience translating social issues into structured coding systems that people can interact with and learn from.

What's next for MediBias

Next, we’d like to expand MediBias by integrating real-world research datasets, refining the inclusivity scoring system, and adding more factors such as race, disability, and age. Long term, we hope MediBias can become an open-source framework that supports researchers, clinicians, and advocates in building a more inclusive healthcare system.