PatientX- Personal Medical Data Marketplace
AI-Powered, Blockchain-Secured Healthcare
Created on 17th December 2025
•
PatientX- Personal Medical Data Marketplace
AI-Powered, Blockchain-Secured Healthcare
The problem PatientX- Personal Medical Data Marketplace solves
The Problem PatientX Solves:
Millions of patients produce valuable medical data through hospitals, wearables, and apps, but:
1.They don’t own or control their data
2.Health records are stored in centralized systems vulnerable to breaches
3.Patients don’t earn any value when companies use their data
4.Researchers struggle to access quality, consent-based medical data due to strict privacy laws
5.Sharing data involves risk, bureaucracy, and no transparency
PatientX solves these problems by creating a system where data ownership becomes patient-centered, secure, and transparent.
What People Can Use PatientX For:
For Patients
1.Store health records securely on a decentralized blockchain
2.Control who accesses their data using smart contracts
3.Share data anonymously through AI-based data anonymization
4.Earn rewards when their anonymized data is used by researchers
5.Verify insights safely using Zero-Knowledge Proofs without exposing identity
6.Prevent misuse by tracking every access in an immutable ledger
For Doctors & Hospitals
1.Access accurate patient data quickly with consent
2.Reduce fraud through transparent, tamper-proof records
3.Enable interoperable health data exchange without paperwork
For Researchers & Healthcare Companies
1.Legally access high-quality, anonymized medical data
2.Speed up medical innovation without violating privacy laws
3.Gain deeper insights validated by AI and ZK-proofs
How It Makes Tasks Easier & Safer
Safer: Decentralization removes single-point data breaches
Faster: Automated smart contract approvals replace manual paperwork
Private: AI-anonymization and zero-knowledge proofs protect identity
Fair: Patients finally earn value from their medical data
Transparent: Every data transaction is traceable and consent-based
Challenges I ran into
1.AI-Based Data Anonymization
Problem: Model missed implicit identifiers in reports.
Solution: Added hybrid neural network + rule-based entity removal + multi-layer validation.
2.Smart Contract Development
Problem: Consent contracts reverted due to logic conflicts.
Solution: Introduced HardHat, unit testing, and contract simulation.
3.OCR Inaccuracy
Problem: Pytesseract failed on low-resolution PDFs and images.
Solution: Pre-processing pipeline (denoising, thresholding, sharpening).
4.Decentralized Storage Handling
Problem: Inconsistent IPFS uploads for multiple formats.
Solution: Standardized file metadata + buffer normalization.
5.Zero-Knowledge Proof Integration
Problem: High computational complexity.
Solution: Used lightweight ZKP primitives + optimized proof generation.
Tracks Applied (2)