Inspiration
During a hospital audit, we realized that while financial systems are well-guarded, clinical fraud often goes undetected. Fake admissions, unauthorized access, and manipulated patient workflows cost hospitals millions — and put lives at risk. That’s where David was born: a tool that thinks like a forensic investigator, built to track patient journeys, audit healthcare workflows, and expose anomalies.
What it does
David is the world’s first Healthcare Forensics Platform, built to analyze HL7 messages and patient events in real time. It captures, classifies, and audits patient movement across departments — revealing red flags and inconsistencies that often go unnoticed.
How we built it
The entire healthcare forensics as a concept is structured into 3 parts. First is a case management platform which allows you to configure and create cases under which the forensics investigation is documented. Now we need Hospital data to simulate workflows, which we cannot get that easily, so we are planning on Level 2 an Open source HIS called ZOIE, that will have inbuilt integration capabilities, and will act as a test harness. The third part is combining both case management and HIS together and we will get David.
Challenges we ran into
HL7 variability across systems made standardization hard. Building a lightweight stack that runs on low-resource environments in Indian hospitals. Designing an intuitive yet powerful UI for investigators.
Accomplishments that we're proud of
Built a working Healthcare Forensics engine from scratch — capable of ingesting, parsing, and auditing real HL7 messages. Designed an end-to-end fraud investigation workflow with real user roles (Investigator, Analyst, Admin, Verifier). Successfully detected simulated fraudulent billing patterns using live patient movement data. Created a beautiful, intuitive UI for investigators — no training manual needed. Engineered David to be lightweight and hospital-friendly, deployable on-premise with minimal infra. Packaged insights into AI-generated reports, complete with timeline, attachments, and audit trail. Developed a custom RBAC system tailored for forensic review, with escalation and verification steps. Most importantly: proved that Healthcare Forensics is not just possible, but essential.
What we learned
HL7 is messy. But with the right parsing and classification techniques, it tells an incredibly rich story. Security and privacy are not just buzzwords — we had to design with HIPAA in mind. Detecting fraud in healthcare isn’t just about numbers, it’s about narratives.
What's next for David
Internationalization. Open source HIS called ZOIE.
Log in or sign up for Devpost to join the conversation.