Sanjeevani - Healthcare Transparency for Every Indian

An AI-powered healthcare billing transparency platform that fights medical fraud by scanning hospital bills and cross-referencing them with 1,995 official CGHS government rates.

Comment

Inspiration

🎯 Inspiration

Medical billing fraud and arbitrary overcharging form a hidden ₹12,000 Crore+ crisis in India's healthcare system. Nearly 80% of private hospital bills contain inflated line items, administrative inaccuracies, or non-compliant rates. These extra charges typically hover between 30% to 400% above fair standards.

Average citizens do not have the time, legal knowledge, or resources to cross-examine complex medical invoices line-by-line. We built Sanjeevani (named after the legendary life-restoring herb) to establish an automated, real-time consumer protection shield that instantly audits bills using official government datasets.

⚙️ What it Does

Sanjeevani is an intuitive ecosystem engineered to catch hospital overcharges and locate lower-cost alternatives instantly:

  • Intelligent Bill Upload: Users snap a photo or drop a PDF of their complex hospital bill.
  • Automated Auditing Engine: The pipeline extracts text dynamically, strips out structural visual noise, and audits the extracted treatments against 1,995 official Central Government Health Scheme (CGHS) rates.
  • Discrepancy Highlighting: If a hospital overcharges a user for a procedure, Sanjeevani flags it clearly, calculating the exact inflation variance.
  • Actionable Claims Report: The platform compiles an organized, downloadable breakdown detailing legal/compliance discrepancies that users can use to officially contest their bills.
  • Jan Aushadhi Store Locator: Integrated via geographic mapping interfaces, it locates the closest generic drug dispensaries to help patients save money on subsequent prescriptions.

🛠️ How We Built It

We engineered a scalable, decoupled architecture to ensure low-latency processing and reliable database operations:

  • Frontend Dashboard: Built utilizing React 19, Vite 6, and Tailwind CSS. Data visualization indicators handle charts cleanly, and mapping configurations hook directly into generic drug discovery networks.
  • Backend Pipeline: Written in Node.js and Express 5 backed by a MongoDB Atlas cluster. Route protection is strictly managed using secure web tokens (JWT) and network sanitation middleware (Helmet).
  • Fuzzy Database Matching: Hospital invoices rarely match government registries letter-for-letter. We deployed Fuse.js inside the application cluster to support text distance threshold scoring, optimizing identification tolerance up to a 99% confidence metric.
  • Dual-Engine OCR Service: A high-availability microservice built using FastAPI and Python. The machine learning execution stack falls back systematically between EasyOCR and Tesseract, entirely isolated inside containerized Hugging Face Spaces Docker clusters.

📊 The Mathematics Behind Sanjeevani

To evaluate bill integrity precisely, our backend applies structural rate calculations. The metric used to flag hospital invoice inflation thresholds is computed using the following LaTeX equation:

$$\text{Overcharge Percentage} = \left( \frac{\text{Hospital Checked Price} - \text{CGHS Compliant Rate}}{\text{CGHS Compliant Rate}} \right) \times 100$$

When the platform identifies line items where $\text{Overcharge Percentage} \ge 20\%$, the processing gateway triggers automated legal compliance alerts on the client interface dashboard.

🚀 Accomplishments That We're Proud Of

  • Achieved an end-to-end multi-page bill processing lifecycle of ~15 seconds.
  • Stabilized OCR scanning extraction boundaries to read complex table formatting with a 98%+ text accuracy metric.
  • Successfully integrated a massive registry of 1,995 independent government compliance entries without experiencing request latency bottlenecks.
  • Built a clean, universally accessible design language engineered specifically for users navigating high-stress medical environments.

🏆 What's Next for Sanjeevani

  • Automated Insurance Claims Integration: Implementing smart contract frameworks to link authenticated audit reports directly to third-party insurance providers for instant dispute resolutions.
  • LLM Legal Assistant: Integrating localized language models trained on national medical healthcare frameworks to draft custom dispute emails automatically.

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Sanjeevani — AI Medical Bill Auditor

Built With

Share this project:

Updates