us/echo

What it does

Echocardiography (TTE) reads a complete transthoracic study. Upload the 30 to 150 clips an echo cart exports, with 2D, M-mode, colour and PW, CW or tissue Doppler mixed together; the run labels every clip by view and mode, scores its quality and runs each measurement only on the clips that support it.

The result is an ASE-structured measurement table with findings per structure: ejection fraction and volumes, LV dimensions and mass, left atrial volume, right ventricular function, strain, regional wall motion, diastolic grade, valve severity, pulmonary hypertension probability, pericardial effusion and cardiomyopathy patterns. A focused section answers the four FoCUS questions from one to five handheld clips.

Intended use

Decision support for the first read of an echo: the fellow or sonographer preparing a study for sign-off, the on-call physician who needs an ejection fraction and an effusion answer at night, and the emergency or ICU doctor scanning with a handheld probe. Each value carries its source clip and beat count.

Who it is for

Cardiologists, echo sonographers, cardiology fellows, emergency and intensive-care physicians and cardiac anaesthesiologists. Developers add the same run to a PACS, an echo reporting system or a heart-failure registry.

Accepted input

• DICOM studies: Ultrasound Multi-frame (cine loops) and Ultrasound Image objects, including colour and spectral Doppler strips, as a study folder, a CD or individual files.
• Vendor measurements: Comprehensive SR measurement objects (TID 5200) in the upload are read as a second source.
• Handheld exports: one to five MP4 or MOV clips (PLAX, PSAX, A4C, subcostal, IVC) for the focused scan and ejection fraction.

Requirements

Requirement	Why
Cine with FrameTime or FrameTimeVector, CineRate and NumberOfFrames	EF, strain and wall motion need frame timing; stills are refused for them
SequenceOfUltrasoundRegions	Calibrates millimetres per region and the velocity axis of each Doppler strip
HeartRate and the ECG trace when present	End-diastole and end-systole come from the ECG or from the LV area curve
Apical 4- and 2-chamber views (3-chamber for strain)	Biplane LV and LA volumes need both apical views; GLS uses all three
3 to 5 beats on each Doppler strip	Envelopes are traced beat by beat and averaged; more beats in atrial fibrillation

Handheld MP4 carries no calibration, so its millimetre values depend on a readable depth scale or a user-entered depth, and the result names the scale source.

Optional context

Height and weight for BSA indexing, sex (ASE normal ranges are sex-specific), age, heart rhythm and the clinical question go in clinical_context. Mark contrast (LVO) studies and prosthetic valves there as well.

The result

One JSON result in sections, with findings, measurements, classifications and a clips[] list giving view, mode, quality and cardiac phases for every clip. A measurement records whether it was computed, read from on-screen calipers or taken from vendor SR. A foreshortened or off-axis clip yields "insufficient view" and its reason.

Section	Content
Views and image quality	View (PLAX, RV inflow, PSAX levels, A4C/A5C/A2C/A3C, subcostal, SSN), mode and quality score per clip
Chamber segmentation	LV endocardium and epicardium, LA, RV and RA masks on every frame, with areas and volumes
LVEF and volumes	Biplane Simpson LVEF with a video-based estimate, EDV and ESV (indexed), beat-to-beat spread, HF category
LV size and mass	IVSd, LVIDd/s, LVPWd, LV mass and index, relative wall thickness and LV geometry
Left atrial volume	Biplane LA volume at end-systole, LAVi and LA diameter
Right ventricular function	TAPSE, fractional area change, S′, RV basal diameter and RV/LV ratio
Global longitudinal strain	GLS (%) and the segmental bull's-eye
Regional wall motion	Per-segment grade (normal, hypokinetic, akinetic, dyskinetic), WMSI and coronary territory
Diastolic function	E, A, E/A, septal and lateral e′, E/e′, TR Vmax, LAVi, diastolic grade and LA pressure estimate
Valve disease	AS (Vmax, mean gradient, AVA by continuity, DVI); MR, AR, TR (vena contracta, PISA EROA, jet); MS gradient; AS screen from 2D-only PLAX/PSAX
Pulmonary hypertension probability	TR Vmax, RVSP, IVC size and collapse (RAP), RA area, PA acceleration time and echo probability
Pericardial effusion	Presence, size (small, moderate, large) and tamponade signs (RA/RV collapse, plethoric IVC)
Cardiomyopathy patterns	HCM versus cardiac amyloidosis versus other LVH, LVOT obstruction, dilated pattern and next-test guidance
Focused cardiac ultrasound	LV function with EF estimate, RV dilatation (RV/LV > 1), effusion with tamponade signs, IVC size and collapse
Draft report	Opt-in (options.report): an English narrative that interprets the result for the reader (findings, impression, limitations); every number is checked against the findings

Standards

• DICOM: SR Echocardiography Procedure Report TID 5200, or the Simplified Echo Procedure Report TID 5300; SEG for chamber masks, GSPS calipers and contours on the source frames, Parametric Map for strain, MP4 overlays and PDF.
• Quantification: ASE 2015 chamber quantification, ASE 2019 comprehensive TTE, ASE 2025 diastolic function and HFpEF update, ASE 2025 right-heart guideline, EACVI/ASE/Industry 2015 strain standardisation, AHA 17-segment model.
• Disease grading: ASE 2017 native valvular regurgitation, EACVI/ASE 2017 aortic stenosis, ESC/ERS 2022 pulmonary hypertension, ESC 2023 cardiomyopathy guideline, ASE 2013 focused cardiac ultrasound.

Grading notes

The HF category follows LVEF: reduced at 40 % or less, mildly reduced at 41–49 %, preserved at 50 % or more. LAVi above 34 mL/m² is enlarged per ASE 2015. Each section records the guideline version its grade came from. The read is built for adult TTE; paediatric and congenital studies are read for ejection fraction only.

Related models

• us/lung and us/vascular complete the breathless-patient work-up at the bedside.
• us/echo-tee reads transoesophageal studies.