AIHC - AYUSH Integrated Health Care

• 

  These are the main features of our project.

Inspiration

We were inspired by a simple problem: doctors often have very little time to understand a patient's complete history, while patients may struggle with paperwork, language, or explaining symptoms. We wanted to make case-taking simpler, faster, and accessible to everyone.

What it does

AIHC – AYUSH Integrated Health Care digitizes the patient journey from registration to consultation. Patients can book appointments and complete an AI pre-consultation in Tamil, English, or Hindi using text or ASR voice input. AI generates adaptive questions based on previous answers, uses a body map for symptoms, analyzes uploaded reports using OCR, and detects potential emergency symptoms to alert doctors and nurses. The platform also supports ABDM consent and DPDP-aligned data protection.

How we built it

We built AIHC with a modern web architecture using React, TypeScript, Vite, Tailwind CSS, FastAPI, and PostgreSQL. The AI layer is designed around adaptive question generation, ASR, emergency detection, and OCR-based report summarization. Role-based access separates patient, doctor, nurse, and lab-technician workflows.

Challenges we ran into

Our biggest challenges were designing AI questions that genuinely adapt to previous answers, supporting Tamil, English, and Hindi voice input, making speech understandable across different accents, and creating an interface that works even for low-literacy and elderly patients. We also had to balance AI capabilities with privacy, security, and clinical safety.

Accomplishments that we're proud of

We are proud of creating a complete digital healthcare workflow rather than just a chatbot. AIHC connects patient case-taking, adaptive AI, ASR, body mapping, OCR report analysis, emergency detection, triage alerts, and role-based hospital workflows into one platform.

What we learned

We learned that healthcare AI is not only about building an intelligent model. Accessibility, human oversight, privacy, consent, security, and seamless workflow integration are equally important. We also learned how multilingual and voice-first design can make technology more useful for people who may struggle with traditional digital interfaces.

What's next for AIHC - AYUSH Integrated Health Care

Next, we aim to improve the AI models with larger multilingual and AYUSH-aware datasets, strengthen ASR and accent understanding, integrate with ABDM/HIS systems, improve emergency detection, and validate the platform with healthcare professionals and real-world clinical workflows.

Built With

• asr
• fastapi
• flan-t5
• ocr
• postgresql
• python
• react
• tailwind-css
• typescript
• vite