AI echocardiography assistants match experts' ejection fractions

AETSA reviewed five studies of four AI echocardiography assistants. It found that their ejection fractions agreed with manual expert measurement, with no systematic error identified.

  • 2 for an organization

    In the studies that AETSA reviewed, ejection fractions from the four systems agreed with expert measurement, with usable nurse scans in 98.8% of cases in one study.

  • Not rated society

    AETSA found no data on use of the systems in Europe and no long-term studies of their diagnostic validity.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudiedreportedclaimed

An independent assessment report supports the benefit.

Compared with

Manual measurement of the ejection fraction by expert echocardiographers or cardiologists with the modified Simpson method. Measured.

The system’s part
Not isolated. The studies compared ejection fractions from the AI systems with manual expert measurement, which tests the systems' accuracy. They did not compare diagnoses or care with and without the systems. In the German study, the software analyzed images that a cardiologist had taken. AETSA describes Caption Guidance and AI TRIO as deep learning and the other two as neural networks.
Not yet checked
LimitsCosts and harmsLearns from dataLater outcome

Download data Cite Suggest a correction

Benefit type
Improved Task Accuracy
System
AI echocardiography assistants: Caption Guidance, Real-Time EF, LVivo Seamless and AI TRIO
Made by
Caption Health (Caption Guidance), GE HealthCare (Real-Time EF), DIA Imaging Analysis (LVivo Seamless) and EchoNous (AI TRIO)
Used by
Hospitals in the five validation studies that AETSA reviewed, in the United States, Germany, Tunisia and Portugal
Where
United States, Germany, Tunisia and Portugal (studies), assessed for Spain
Stage
In a pilot
Date
, when the source was published
Kind
Aggregate result
Broad benefit
Estimating the Quantifiable Characteristics of Products, Events, or Information, Training and Teaching Others
Collections
Scrutiny and findings
Last revised
1 November 2026

Sources

  1. Assessment report, by authors with no tie to the system.

    AETSA, Evaluación de Tecnologías Sanitarias de Andalucía (Maesa Márquez JM, Isabel Gómez R). Sistema de identificación de enfermedades cardiacas basado en ecografía e inteligencia artificial. Ficha de evaluación de tecnologías sanitarias nuevas y emergentes, RedETS and Ministerio de Sanidad, Sevilla, 2024. Puntos clave, p. 53.

    Los trabajos incluidos determinan la existencia de concordancia entre la FEVI determinada por los cuatro sistemas de IA y las obtenidas manualmente por el método de referencia, sin identificarse ningún error sistemático.

Details

AI system

Workflow

Nurses or residents without echocardiography experience scanned patients, after one hour of training in one study and one day in another. The software guided the probe and judged image quality, except LVivo Seamless, which analyzed stored images. The software computed the ejection fraction automatically from the apical views.

Follow-up

Conditions for replication

The software must run on compatible scanners, such as the Terason uSmart 3200T Plus for Caption Guidance and the Venue for Real-Time EF. Expert cardiologists must supervise use, since AETSA writes that the results should not replace a specialist's diagnosis.

Entry details

Sector

Healthcare (cardiology).

Purpose

Caption Health, GE HealthCare, DIA Imaging Analysis and EchoNous built Caption Guidance, Real-Time EF, LVivo Seamless and AI TRIO to help users without expert training perform echocardiography. The systems also compute the left ventricular ejection fraction automatically.

31 more questions not yet checked

model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). AI echocardiography assistants match experts' ejection fractions (AIBD-1030). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-1030.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)

BibTeX Data Field definitions