Doctor Who?

Main symptom screen
Example recommendation

Inspiration

We were inspired by the everyday confusion people feel when they do not know which kind of doctor to see first. A vague symptom like “I can’t see clearly” or “my foot hurts” should not force someone into a generic family physician path. We wanted to create a friendly guide that helps people move from uncertainty to the right kind of specialist.

What it does

Doctor Who? is a static web app that asks a user about symptoms and location, then recommends the most likely specialty and nearby doctors. It uses semantic phrases, keywords, and location parsing to suggest a doctor specialty such as ophthalmologist, ENT, orthopedic, gastroenterology, or primary care.

How we built it

We built the app with HTML, CSS, and JavaScript for the front end, with a Python local HTTP server acting as a lightweight prediction API. The front-end UI is rendered through the HTML and styling files, while the specialty scoring and rendering logic lives in the JavaScript. The server returns JSON specialty predictions, confidence, reason traces, and AI-style assessment messages.

Challenges we ran into

The biggest challenge was moving beyond brittle keyword matching. We needed the app to recognize deeper semantic patterns like “blurry vision,” “I can’t hear,” and “my foot hurts” and map them to the correct specialty. Another challenge was making sure generic primary-care suggestions were not ranked above clear specialty evidence, while still keeping the app accessible and reassuring.

Accomplishments that we're proud of

We built a location-aware doctor recommendation flow, added semantic specialty ranking, created an expanded specialty vocabulary, and made the UI cleaner and more human-friendly. We also pushed the project toward a smoother experience by improving the match cards, doctor recommendation layout, and symptom-to-specialty reasoning.

What we learned

We learned that healthcare recommendations need to be both emotionally reassuring and technically grounded. A good recommendation engine should explain its reasoning, avoid overusing generic physician suggestions, and prefer real specialty intent whenever symptoms clearly point to one.

What's next for Doctor Who?

The next step is to connect the app to a real backend or medical API, expand the doctor and specialty dataset, and improve the model so it can make stronger, more general symptom-specialist matches across more body systems and languages.