Inspiration
What it does
How we built it
Challenges we ran into
Accomplishments that we're proud of## 🌟 Inspiration
Access to specialized dermatological care is a global challenge. With skin cancer rates rising, early detection and precise follow-up are more critical than ever. Many general practitioners lack the tools to accurately track the subtle evolution of lesions over time. We built DermoPro AI to bridge this gap—providing every physician with a high-tech clinical workspace powered by the world's most advanced multimodal AI.
🚀 What it does
DermoPro AI is a comprehensive mobile ecosystem for dermatologists and general practitioners:
- Secure HUD Portal: A futuristic, laser-scanning interface for physician authentication.
- Dynamic Patient Management: Full CRUD capabilities to manage records, search patient history, and monitor risk levels in real-time.
- Multimodal AI Scan: Uses Gemini 3 to perform instant clinical assessments of lesions based on the ABCDE criteria.
- Evolution Mode (The Game Changer): Gemini 3 analyzes two images of the same lesion from different dates side-by-side, detecting minute changes in size, color, and borders that the human eye might miss.
- AI Medical Scribe: Converts informal, voice-dictated notes into professional, structured clinical observations instantly.
- AI Research Lab: A dedicated hub for technical differential diagnosis and an expert chatbot to consult on rare cases or medical protocols.
- Professional PDF Export: Generates clinical-grade reports ready for hospital records or patient sharing.
🛠️ How we built it
- Frontend: Built with Flutter for a high-performance, cross-platform experience.
- Backend: Firebase (Authentication & Cloud Firestore) ensures secure, real-time data synchronization.
- AI Engine: The core is the Gemini 3 API (gemini-3-flash-preview). We leveraged its multimodal capabilities to process both high-resolution images and complex medical text simultaneously.
- Reporting: Integrated specialized PDF and Printing libraries to transform AI reasoning into official medical documentation.
🧠 Challenges we ran into
One of the biggest challenges was ensuring that the AI’s reasoning followed strict dermatological protocols. We spent significant time on Prompt Engineering to make sure Gemini 3 prioritized clinical semiology over generic descriptions. Another hurdle was handling the multimodal data flow—syncing historical "baseline" images with current scans to provide a meaningful "Evolution Delta."
🎉 Accomplishments that we're proud of
We are incredibly proud of the Evolution Mode. Seeing the AI accurately identify that a lesion's borders have become more irregular or that its diameter has increased by a specific percentage over six months is a massive step forward for digital health.
📖 What we learned
We discovered the true power of Gemini 3’s low-latency reasoning. It’s not just about "seeing" an image; it’s about "understanding" the clinical context and the temporal changes between two data points. We also learned how to build a highly secure medical environment while maintaining a fluid, modern UI.
🔮 What's next for DermoPro AI
We plan to integrate 3D body mapping to visualize lesion distribution across the entire body and implement a "Lab OCR" feature to automatically scan and interpret biopsy results from physical papers.
Log in or sign up for Devpost to join the conversation.