# Echocardiography (TTE) (us/echo)

> Full TTE read: views, LVEF, chambers, diastology, valves and effusion as ASE measurements.

Upload a transthoracic echo study (the 30–150 clips an echo cart exports) or one to five handheld MP4 clips. One run labels every clip by view and mode with a confidence gate, estimates LVEF from the apical cine clips with two video networks, and returns EDV, ESV (indexed when height and weight are given), the LVEF category, the ASE 2015 grade of LV systolic function and the FoCUS LV-function answer. Every value names its source clips. A draft report is available as an opt-in add-on.

Model page: https://medrun.ai/models/us/echo
Status: beta · version 1.0.0 · viewer: cine

## Input

Whole TTE study (DICOM folder, CD or zip; colour and Doppler clips may stay in) or 1–5 handheld clips (MP4/MOV) with apical 4- and 2-chamber views. Add height and weight for indexed volumes.

Accepted: DICOM files or folder, CD export (DICOMDIR), zip, MP4 / MOV / AVI clips, PNG / JPEG stills.

## Pricing

- `us/echo/interpret`: $0.06 up to 1,000 frames · $0.15 up to 8,000 frames · $0.25 above 8,000 frames per request
- Draft report (opt-in, `options.report`): +$0.02 per run

## Result sections

- `#views` Views and image quality (available): View (PLAX, PSAX, A2C, A3C, A4C, A5C, subcostal, suprasternal) and mode (2D, colour, spectral, M-mode) per clip; clips below the view-confidence gate are named and not measured.
- `#segment-chambers` Chamber segmentation (planned): LV endocardium and epicardium, LA, RV and RA masks on every frame, with areas and volumes.
- `#lvef` LVEF and volumes (available): LVEF from the apical cine clips (mean of two video networks, recalibrated), EDV and ESV (indexed by BSA), sex-specific ASE ranges, LVEF category and the ASE grade of LV systolic function.
- `#lv-size` LV size and mass (planned): IVSd, LVIDd/s, LVPWd, LV mass and index, relative wall thickness and LV geometry.
- `#la-volume` Left atrial volume (planned): Biplane LA volume at end-systole, LAVi and LA diameter.
- `#rv-function` Right ventricular function (planned): TAPSE, fractional area change, S′, RV basal diameter and RV/LV ratio.
- `#lv-strain` Global longitudinal strain (planned): GLS and the segmental bull's-eye.
- `#wall-motion` Regional wall motion (planned): Per-segment grade, WMSI and coronary territory (AHA 17-segment model).
- `#diastolic` Diastolic function (planned): E, A, E/A, e′, E/e′, TR Vmax, LAVi, diastolic grade and LA pressure estimate.
- `#valves` Valve disease (planned): Aortic stenosis and regurgitation grading from Doppler and 2D.
- `#pulmonary-hypertension` Pulmonary hypertension probability (planned): TR Vmax, RVSP, IVC size and collapse, RA area and the ESC/ERS echo probability.
- `#pericardial-effusion` Pericardial effusion (planned): Presence, size and tamponade signs.
- `#cardiomyopathy` Cardiomyopathy patterns (planned): HCM versus cardiac amyloidosis versus other LVH, LVOT obstruction and dilated pattern.
- `#focused` Focused cardiac ultrasound (available): FoCUS LV function (normal, reduced, severely reduced) with the EF estimate, and which FoCUS views the upload holds; RV, effusion and IVC answers join with their sections.
- `#report` Draft report (available): Opt-in draft echo report (findings, impression, limitations) written from the structured result; every number is validated against it.

## Endpoint `us/echo/interpret` (run)

The full read of the upload: every clip is classified by view and mode, routed to the sections that can use it, and the result contains every section the input supports, the FoCUS answers included.

Docs: https://medrun.ai/models/us/echo/api · OpenAPI: https://api.medrun.ai/v1/models/us/echo/interpret/openapi.json

### Input schema

- `files` (array; required): The echo study: DICOM files, a folder, a CD export (DICOMDIR) or a zip, or 1–5 handheld clips (MP4, MOV, AVI). `file_…` ids from /v1/uploads or https URLs.
- `clinical_context` (object): Optional. Sex selects the ASE normal ranges; height and weight give BSA-indexed volumes. Never names, IDs or dates.
  - `age_years` (integer): Age (years)
  - `sex` (string; one of female, male, other, unknown): Sex
  - `height_cm` (number): Height (cm)
  - `weight_kg` (number): Weight (kg)
  - `indication` (string): For example "dyspnoea" or "heart failure follow-up".
- `options` (object): Options
  - `report` (boolean; default false): Adds a draft echo report written from the structured result (English). +$0.02 per run.
  - `return` (array; default ["json"]): JSON is always returned, with a view overview image.

### Example (cURL)

```bash
response=$(curl --request POST \
  --url https://api.medrun.ai/v1/run/us/echo/interpret \
  --header "Authorization: Bearer $MEDRUN_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "input": {
    "files": [
      "file_01K6R4X2Q9Y7T3V5W8Z0A1B2C3"
    ],
    "clinical_context": {
      "sex": "female",
      "height_cm": 165,
      "weight_kg": 70
    },
    "options": {
      "report": false
    }
  }
}')
REQUEST_ID=$(echo "$response" | jq -r .id)
```

## Output

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