edc-retinopathy

Add a diabetic retinopathy screening form and REST API for integrating with a fundus camera in a clinicedc project.

A clinician creates an EyeExamRegister instance in the EDC before the exam. The camera software (fundus-camera-watchdog) resolves the subject, uploads eye images, DICOM files and reports, then checks status.

Requires django>=5.2,<6.1 and djangorestframework>=3.16.0

REST API

All endpoints use TokenAuthentication.

GET /api/retinopathy/ping/

Health check. Returns {"status": "ok"}.

POST /api/retinopathy/resolve/

Confirm a EyeExamRegister exists for a subject. Payload:

{"subject_identifier": "105-10-0001-2"}

Returns 200 with eye_exam_register_id and uploaded list, or 404 if no session exists.

POST /api/retinopathy/<subject_identifier>/<file_type>/

Upload a file. Multipart form with file, capture_datetime, and optional checksum (SHA-256). Valid file types:

• left, right -- eye images (JPEG/PNG)
• left_dicom, right_dicom -- DICOM files
• left_report, right_report, report -- reports (PDF/HTML)

Multiple files per type are accepted. Files are stored with their original filenames under <storage_dir>/images/<session_pk>/.

GET /api/retinopathy/<subject_identifier>/status/

Session status: uploaded types, missing types, and whether the session is complete.

Models

EyeExamRegister

Created by the clinician in the EDC. Links to RegisteredSubject and stores contra-indication screening fields, report type (combined / per-eye), and device ID.

SessionFile

One record per uploaded file. Tracks file type, original filename, stored path, content type, size, SHA-256 checksum, and capture timestamp.

Settings

EDC_RETINOPATHY_STORAGE_DIR

Base path for file storage. Must contain an images/ subdirectory.

EDC_RETINOPATHY_MAX_FILE_SIZE_MB

Maximum upload size in MB (default: 10).

EDC_RETINOPATHY_VISIT_DATETIME_FILTER

Datetime filter for Call List. If set, filters the list of subjects to those who attended AFTER the given datetime.

See also https://edc-retinopathy.readthedocs.io/en/latest/