Verisure

Demo

About

The problem it solves For many patients, insurance claim delays aren’t just frustrating—they’re life-threatening. People facing urgent medical conditions, including terminal illnesses, are often left in limbo, unsure whether they can afford life-saving treatments while waiting for insurers to process claims. These long wait times create immense emotional, physical, and financial stress, forcing patients and families into impossible choices. Meanwhile, insurance companies face exploitation by bad actors looking to abuse the system with fraudulent claims, increasing the burden on the patient with tighter restrictions and increased insurance premiums. These increases make healthcare more out of reach for the average person. The result? A vicious cycle where trust erodes, costs rise, and patients suffer. VeriSure aims to cut out the Medical Claims Clearinghouse middlemen and provide a faster, decentralized, and secure way to process medical claims. User Interaction and Data Flow ### **Hospital submits claim (PDF) → AI extracts data.** - **User Action: Hospital staff logs into the VerifiSure web app and uploads a patient insurance claim document (PDF).** - **Backend Processing: AI (OCR + NLP) extracts relevant claim details: patient ID, policy number, claim amount, and diagnosis.** - **Storage: Extracted data is stored in a structured format (JSON) for processing.** ### **zkVM generates ZK proof.** - **Backend Processing: zkVM guest program generates a ZK proof that verifies:** - **The hospital is registered.** - **The claim form data consists of all the required fields filled out.** - **Future: The patient’s policy covers the requested treatment.** - **Result: The ZK proof is submitted to the smart contract for verification.** ### **EigenLayer operators either approve payout or flags claim for dispute.** - **User Action: Insurers log into the web app to review flagged claims.** - **Backend Processing:** - **If the ZK proof verifies the claim, the smart contract automatically approves the payout.** - **If fraud is detected, the smart contract flags the claim and allows the insurer to dispute it.** - **Result:** - **Valid claims: Smart contract initiates payout to the hospital’s wallet through escrow account funds.** - **Flagged claims: Insurer must submit a counter ZK proof to dispute the fraud flag.** The project architecture and development process **The VerifiSure product is a decentralized web-based application designed for seamless claim processing and fraud detection.** ### **User Interface (Frontend - React/Next.js)** - **Claim Submission Page (Hospitals) → Upload PDFs, view processing status.** - **Fraud Dashboard (Insurers & Auditors) → View claim risk scores, challenge fraudulent claims.** - **Claim Approval Page (Insurers) → Accept or dispute claims via ZK proofs.** ### **Backend & Processing** - **FastAPI (Python) for AI OCR + NLP processing.** - **Solidity Smart Contracts (AVS & Payouts) on EigenLayer.** - **Boundless zkVM for zero-knowledge proof verification.** ### **On-Chain Components** - **AVS Contract (Othentic) → Tracks fraud scores & claim history.** - **Gaia ZK Proof Verifier → Validates claim authenticity & disputes.** - **Payout Smart Contract → Manages claim payouts & disputes.** Product Integrations ### Why Othentic? VeriSure benefits from the Performer, Attester, and Aggregator Node architecture of Othentic. Performer nodes can be static, randomly chosen, election-based or a staked mechanism In VeriSure, the Performer node will handle the sensitive information, ensuring HIPPA compliance by computing a zero-knowledge proof for the validity of the claim off-chain ### Why Gaia? VeriSure integrates Gaia Nodes to enhance claim validation using historical data. By leveraging Gaia’s AI-driven CMS 1500 claim form analysis, VeriSure ensures that claims are pre-validated off-chain before ZK attestation to imporve accuracy and assist clearinghouses in determining reasons why the claim mighht be incomplete or inaccurate. - Gaia Nodes analyze past claim data to detect patterns that could lead to rejection using a small dataset of good and bad claims to base decisions off of. - Ensures claims are scrubbed for completeness before processing, reducing unnecessary denials. In VeriSure, the Gaia Node acts as an intelligent pre-validator to minimize errors and enhance fraud detection without compromising patient privacy. ### Why zkVM? The idea is simple: low cost and high scalability Generally, zero-knowledge proofs allow us to verify protected health information (PHI) across a community of decentralized nodes without needing to expose PHI to them. With zk proofs, we can also enable the verifiability Key differentiators and uniqueness of the project AI-Powered Claim Scrubbing – Uses Gaia Nodes to analyze past data and predict claim rejection risks before processing. Zero-Knowledge Proofs for Compliance – Verifies claims without exposing sensitive patient data, ensuring HIPAA compliance and fraud-proof validation. Decentralized AVS on EigenLayer – Removes the need for medical clearinghouses, reducing costs and increasing claim processing speed. RISC Zero zkVM Auditing – Ensures mathematical proof of claim legitimacy, reducing disputes and increasing transparency for insurers, hospitals, and regulators. With VeriSure, claim approvals are faster, fraud detection is smarter, and patient privacy is always protected. Trade-offs and shortcuts while building Future Steps: Expand Cross-Verification Sources Full EigenLayer Integration Real-Time Patient Dashboards Multi-Insurer Proof Marketplace Open Standard for Claim Proofs Compliance as Code (ZK Circuits) Additional Features N/A