ScanAhead

Inspiration

Delayed medical care is a common frustration across Canada, with minor injuries or conditions often going unchecked because navigating clinics, referrals, and waiting rooms takes too long. Patients have symptoms but lack an easy way to communicate them, while doctors spend hours triaging cases that could have been clarified upfront. We were inspired to streamline this process with ScanAhead, where users can describe their symptoms and upload 3D scans, connecting them to the right doctor faster and give clarity on care without the delays.

What it does

Patient Side

Users create an account and submit a case form describing their condition, both visually and verbally through 3D scanning affected areas, recording a voice memo, and specifying a preferred appointment or provider.

Doctor Side

Doctors log in to see cases submitted to them, with a neat form displaying each patient's clinical report and an AI summary of the overall information provided. Doctors can review the case in advance, come prepared for check-ups, and provide a more detailed and evaluated prognosis that helps patients determine their next steps.

How we built it

We built ScanAhead with a hybrid stack optimized for efficient medical case submission and review.

Frontend

Built with React and TypeScript for user interface and animation, and Vite for faster build. Three.js (through React Three Fiber) with Drei for rendering the 3D point cloud model and interaction.

Backend

Built the database with Supabase for secure user authentication, real-time storage of patient data, and case submissions corresponding to patients with appointed doctor, appointment date, 3D scanning, photo imaging, voice recordings, and an AI summary of the case.

KIRIEngine API

We use KIRI to convert uploaded images and videos into a 3D Model

Gemini API

We use the Gemini to generate a structured clinical summary for each patient, combining written symptoms, voice recordings, and images or 3D scans of the affected area into coherent clinical insights for Doctors to review.

ElevenLabs API

We integrated the ElevenLabs API for speech-to-text in the patient form, allowing users to leave voice memos about their condition while responding to guided clinical questions.

Challenges we ran into

We had a clear vision on the problem we wanted to solve, but shaping it into a project that was both impactful and achievable within a single day took a lot of time planning. On the technical side, integrating multiple systems such as features, APIs, and the database into one cohesive application presented real challenges especially during troubleshooting and debugging.

Accomplishments that we're proud of

Have a functioning and cool project

What we learned

Copilot is amazing!

What's next for ScanAhead

Even after finalizing the project, we continued to discover gaps and flaws in our proposal as we built it. If time allows, maybe we could solve those gaps and actualising it as a real world application.

Built With

• elevenlabs
• gemini
• kiriengine
• react
• supabase
• typescript
• vite