04 /
NEXARAD
Evidence-Linked Radiology
Evidence-linked imaging workflows with strict non-clinical demo boundaries.
- Role
- Founder & Engineer
- Status
- Acquired · research foundation
- Verified proof
- Synthetic DICOM demo · PHI disabled · external AI off by default
Demo / Research / Not for Clinical Use
From constraint to evidence
The work, end to end.
Problem
Medical imaging demos often skip the hard parts: PHI boundaries, DICOM plumbing, tenant isolation, object storage, report provenance, and clear labeling when something is research-only rather than clinical software.
Approach
I built a local stack with web, API, OHIF, Orthanc, MinIO, Alembic migrations, seeded non-PHI studies, synthetic upload smoke tests, demo/research safety defaults, and a rule that browser clients only call the NexaRad API.
Outcome
The founder-ready walkthrough can boot the full local demo, seed visible research-only studies and draft reports, rehearse uploads, and keep AI providers and PHI disabled by default.
Overview
What shipped
NexaRad is a production-shaped foundation for radiology workflow software: backend-owned auth, tenant isolation, local DICOM services, object storage, structured findings, verifier-gated reports, and AWS-first infrastructure boundaries.
Built with
- Next.js
- FastAPI
- DICOM
- OHIF
- Orthanc
- MinIO
Evidence
Inside the system
