💡 Inspiration

When people feel sick, they usually turn to internet search engines — a habit that often leads to "cyberchondria" 😰, unnecessary anxiety caused by generic, worst-case-scenario results. Not everyone has immediate access to a doctor for minor triage, and navigating emergency first-aid advice online can be dangerously slow. ⏳

We built MedIntelli to bridge the gap between initial discomfort and professional care — a safe, instant, and highly accessible Clinical AI Assistant that feels like a real conversation. 🩺


⚙️ What It Does

MedIntelli acts as your intelligent first point of contact for health concerns. Users input their age, gender, and current medications (to check for critical drug interactions 💊). Instead of just typing, users can:

  • 🎙️ Voice Input — describe symptoms naturally using speech
  • 🫀 3D Body Map — tap directly where it hurts on an interactive body diagram

In seconds, our AI analyzes the structured data and provides safe, actionable guidance. Beyond triage, the platform features:

  • 🧩 Diagnosis Quiz — interactive symptom checking
  • 📈 Symptom Tracker — long-term log patients can share with their real doctors
  • 🚨 Emergency SOS Guide — bypasses AI entirely to deliver instant, visual first-aid for strokes, CPR, choking, and more
  • 🌍 Multi-language Support — making healthcare knowledge accessible globally

🛠️ How We Built It

We built MedIntelli with a clean glassmorphism UI — desktop & mobile responsive — using vanilla HTML5 and CSS3.

  • 🖥️ Frontend: Vanilla JavaScript handles dynamic severity sliders, interactive SVG Body Map, and Web Speech API for voice-to-text
  • 🔐 Auth & Database: Firebase (Authentication + Firestore) for secure login, Medical ID profiles, and persistent consultation history across devices
  • 📊 Performance & Analytics: Firebase Analytics and Performance monitoring for fast load times — especially critical when users are stressed and need rapid answers

🚧 Challenges We Ran Into

  • 🗂️ Structuring AI input — pairing free-text symptoms with quantitative data (pain scale 1–10, SVG body coordinates) required careful prompt parsing
  • 🔄 Session management — preventing race conditions when fetching Medical ID from Firestore required robust async auth listeners and loading states
  • 🛡️ AI guardrails — ensuring the AI never confidently gives definitive diagnoses, but always offers safe triage and responsible guidance

🏆 Accomplishments We're Proud Of

  • ✨ A premium UI/UX that feels polished and professional
  • 🎙️ Native voice input + interactive Body Map = a far more accessible experience, especially for elderly users
  • 🚨 The Emergency SOS Module — we recognized that relying on an AI during an active heart attack is dangerous, so we built a dedicated hard-coded module that delivers pure visual first-aid in one click

📚 What We Learned

  • 🏗️ Building complex, stateful apps without heavy frameworks
  • 🔥 Deep Firebase integration — real-time data persistence and secure Firestore rules
  • 🧬 Clinical Prompt Engineering — how to make AI act conservatively and safely when handling human health data

🚀 What's Next for MedIntelli

  • 📸 Multimodal AI Vision — upload a photo of a rash, wound, or swelling for AI visual analysis alongside symptoms
  • 👨‍⚕️ Doctor Portal — generate a secure PDF or encrypted link of your Symptom Tracker to send directly to your healthcare provider
  • 📱 Native iOS/Android App — bringing Emergency SOS guides fully offline for when you need them most

⚠️ MedIntelli is an AI-assisted triage tool and does not replace professional medical advice. Always consult a qualified healthcare provider for diagnosis and treatment.

Built With

Share this project:

Updates