Documentation — ClinexaOS

Quick Start

ClinexaOS provides a RESTful API for submitting medical imaging data and receiving structured AI analysis outputs. The API is available at https://api.clinexaos.com/v2.

You'll need an API key from your institution's ClinexaOS account dashboard to authenticate requests.

Authentication

All API requests must include your API key in the Authorization header using Bearer token format:

# Include in all requests
Authorization: Bearer cx_live_xxxxxxxxxxxxxxxxxxxx

# Example with curl
curl -X POST https://api.clinexaos.com/v2/analyze \
  -H "Authorization: Bearer cx_live_xxxx" \
  -H "Content-Type: application/json" \
  -d @payload.json
          

Your First API Call

Submit a DICOM file URL for analysis. ClinexaOS will return a structured JSON response containing findings, confidence scores, and formatted report text.

# Request
POST /v2/analyze

{
  "study_url": "https://your-pacs.hospital.org/studies/CT_0094.dcm",
  "modality": "CT",
  "body_part": "CHEST",
  "specialty_context": "PULMONOLOGY",
  "output_format": "structured_report"
}
          

# Response (abbreviated)
{
  "study_id": "cx_study_8f3a2b91",
  "status": "complete",
  "inference_time_ms": 1842,
  "confidence": 0.974,
  "findings": [...],
  "report": {
    "impression": "No acute cardiopulmonary findings...",
    "structured_data": {...}
  }
}
          

Analysis Endpoint

The POST /v2/analyze endpoint accepts the following parameters:

study_url (required) — Presigned URL or accessible URI of the DICOM study or image file
modality (required) — Imaging modality: CT, MRI, CXR, US, PET, NM
body_part (required) — Anatomical region: CHEST, HEAD, ABDOMEN, SPINE, MSK, PELVIS
specialty_context (optional) — Clinical specialty for context injection: RADIOLOGY, CARDIOLOGY, PULMONOLOGY, ONCOLOGY, NEUROLOGY
output_format (optional) — Response format: structured_report, raw_findings, fhir_diagnostic_report
patient_context (optional) — Anonymised patient metadata to improve contextualisation

DICOM & PACS Integration

ClinexaOS supports direct DICOM integration via WADO-RS and WADO-URI protocols. For PACS-integrated deployments, we provide a lightweight DICOM router that can be deployed within your network perimeter, eliminating the need to transmit imaging data over the public internet.

On-premise option: Enterprise customers can request deployment of the ClinexaOS inference node within their own network. Contact your account team for architecture details.

HL7 FHIR Output

ClinexaOS can return analysis results formatted as FHIR R4 DiagnosticReport resources, enabling direct ingestion into FHIR-compatible EMR systems without transformation middleware:

# Request FHIR output
{
  ...,
  "output_format": "fhir_diagnostic_report"
}

# Returns a FHIR R4 DiagnosticReport resource
{
  "resourceType": "DiagnosticReport",
  "status": "final",
  "category": [{ "coding": [{ "code": "RAD" }] }],
  ...
}
          

Error Codes

400 — Bad Request: missing required parameter or invalid payload format
401 — Unauthorized: invalid or expired API key
403 — Forbidden: modality or specialty not included in your licence
413 — Payload Too Large: DICOM file exceeds 500MB single-study limit
422 — Unprocessable: file format not supported or corrupted study
429 — Rate Limited: see rate limit documentation
503 — Service Unavailable: see API Status page

SDKs & Libraries

Official SDKs are available for the following languages:

Python: pip install clinexa-python
Node.js: npm install @clinexa/sdk
Java: Available via Maven — see documentation for coordinates
C# / .NET: Available via NuGet — dotnet add package ClinexaOS.SDK

For questions not covered here, visit Support or email info@clinexaos.com.

Build with ClinexaOS.