Local-first healthcare AI: clinical NER & HIPAA PII de-identification that runs 100% on-device. 2,200+ medical models, 21 languages, Apple MLX + Python, no cloud, no patient data leaving your network. Apache-2.0
The Helios FHIR Server is an implementation of the HL7® FHIR® standard, built in Rust for high performance and optimized for clinical analytics workloads.
A pure-Rust bridge between openEHR and two interoperability targets: HL7 FHIR via the FHIRconnect specification, and the OMOP Common Data Model via OMOCL, over the openEHR ITS-REST API.
The shared SMART and OAuth 2.0 resource-server crate for every FerroHEALTH product: token validation against the authorization server's keys, SMART v2 scopes and the patient compartment, Basic and client-credentials machine auth, and the audit line.
Pure Go DICOM library and CLI — no CGO. File I/O, pixel decoding (JPEG, JPEG-LS, JPEG Lossless, RLE), and full DIMSE networking: C-STORE, C-FIND, C-MOVE, C-GET, N-DIMSE, UPS. 169 storage SOP classes, 37 transfer syntaxes, DICOMDIR, structured reports, PS3.15 de-identification. The Go equivalent of pydicom + pynetdicom.
Decode, search, validate, and crosswalk US medical codes — ICD-10-CM, ICD-10-PCS, HCPCS Level II, RxNorm — over a bundled offline index via MCP. STDIO or Streamable HTTP.
A pure Node.js monorepo containing HL7 client and server packages. Easily send properly formatted HL7 messages to a broker, or parse and extract specific segments from incoming messages.