# Musculoskeletal X-ray (xr/msk)

> Bone and joint radiograph read: fractures, dislocations, bone age, osteoarthritis and alignment.

Upload an extremity, joint, hand or pelvis radiograph study with all its views, as DICOM or PNG/JPEG. One run routes every image by body part and projection, groups the views of each region and side into one exam, and looks for fractures on all of them together: each fracture is boxed on the view that shows it, placed by region and side, and every exam gets a positive or negative verdict. A child's left hand adds a Greulich & Pyle bone age with its z-score against chronological age. Fracture verdicts cover children's wrists (age under 18 from the DICOM or the clinical context) in this version; other regions and adults are reported as not assessed, never as negative, until the detector meets its accuracy bar there.

Model page: https://medrun.ai/models/xr/msk
Status: beta · version 1.0.0 · viewer: image

## Input

One MSK radiograph study with every view (AP, lateral, oblique), as DICOM files, a folder, a CD export or a zip; PNG/JPEG exports work too. For bone age: the child's left hand PA, with age and sex in the DICOM or the clinical context.

Accepted: DICOM files or folder, CD export (DICOMDIR), zip, PNG / JPEG radiographs.

## Pricing

- `xr/msk/interpret`: $0.04 per request
- Draft report (opt-in, `options.report`): +$0.02 per run

## Result sections

- `#fractures` Fractures and dislocations (available): Fracture boxes on every view with region and side, and a positive or negative verdict per exam (region and side). Covered in 1.0: children's wrists. Other regions, adults, patients of unknown age and views the router cannot place with confidence are reported as not assessed. AO/OTA bone and segment codes and Salter–Harris typing join in a later version.
- `#bone-age` Bone age (available): Greulich & Pyle-equivalent bone age in months from a child's left hand PA, with the z-score against chronological age (Brush Foundation SDs).
- `#knee-oa` Knee osteoarthritis (planned): Kellgren–Lawrence grade per knee, OARSI joint-space narrowing and osteophyte grades.
- `#ddh` Hip dysplasia (DDH) (planned): Hilgenreiner and Perkin lines, acetabular index and IHDI grade on an infant AP pelvis.
- `#bone-tumors` Bone lesions (planned): Bone lesion box with benign versus malignant appearance.
- `#report` Draft report (available): Opt-in draft report (findings, impression, limitations) written from the structured result; every number is validated against it.

## Endpoint `xr/msk/interpret` (run)

The full read of the MSK study: every radiograph is routed, the views of each region and side are read together, and the result contains all sections the input supports.

Docs: https://medrun.ai/models/xr/msk/api · OpenAPI: https://api.medrun.ai/v1/models/xr/msk/interpret/openapi.json

### Input schema

- `files` (array; required): The MSK radiograph study: DICOM files, a folder, a CD export or a zip; PNG/JPEG radiographs are read too. `file_…` ids from /v1/uploads or https URLs.
- `clinical_context` (object): Optional. Age and sex are used for bone age when the files do not carry them; never names, IDs or dates.
  - `age_years` (number): Age (years)
  - `sex` (string; one of female, male, other, unknown): Sex
  - `indication` (string): For example "fall on outstretched hand" or "short stature".
- `options` (object): Options
  - `report` (boolean; default false): Adds a draft MSK radiograph report written from the structured result (English). +$0.02 per run.
  - `return` (array; default ["json","dicom-sr"]): JSON is always returned. DICOM objects reference your original UIDs.

### Example (cURL)

```bash
response=$(curl --request POST \
  --url https://api.medrun.ai/v1/run/xr/msk/interpret \
  --header "Authorization: Bearer $MEDRUN_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "input": {
    "files": [
      "file_01K6R4X2Q9Y7T3V5W8Z0A1B2C3"
    ],
    "options": {
      "report": false
    }
  }
}')
REQUEST_ID=$(echo "$response" | jq -r .id)
```

## Output

The MedRun result envelope: https://medrun.ai/docs/result-envelope