Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Jul 31;26(8):1402-1405.
doi: 10.1093/ehjci/jeaf147.

Deep learning-enabled echocardiographic assessment of biventricular ejection fractions: the dual-task QUEST-EF model

Ádám Szijártó  1 , Béla Merkely  1 , Attila Kovács  1   2 , Márton Tokodi  1   2 ; QUEST-EF Investigators
Collaborators, Affiliations

Deep learning-enabled echocardiographic assessment of biventricular ejection fractions: the dual-task QUEST-EF model

Ádám Szijártó et al. Eur Heart J Cardiovasc Imaging. .
No abstract available

PubMed Disclaimer

Conflict of interest statement

Conflict of interest: Á. Szijártó, Á. Szeier, and B. Magyar are employees of Argus Cognitive, Inc. and receive financial compensation for their work (unrelated to the content of the present manuscript). A.C. has received educational grants, personal fees, and non-financial support from Janssen-Cilag, Ltd. (unrelated to the content of the present manuscript). B.K.L. and A.F. report personal fees from Argus Cognitive, Inc. (unrelated to the content of the present manuscript). E.S. has received personal fees from GE Healthcare and 123sonography and is an employee and a shareholder of AstraZeneca (all unrelated to the content of the present manuscript). M. Takeuchi has received grants from Philips Medical Systems (unrelated to the content of the present manuscript). B. Merkely has received grants from Boston Scientific and Medtronic and personal fees from Biotronic, Abbott, AstraZeneca, Novartis, and Boehringer Ingelheim (all unrelated to the content of the present manuscript). R.M.L. is in the speaker’s bureau and receives grant support from Philips Medical Systems (unrelated to the content of the present manuscript). A.K. reports personal fees from Argus Cognitive, Inc. and CardioSight, Inc. (unrelated to the content of the present manuscript). M. Tokodi reports consulting fees from CardioSight, Inc. (unrelated to the content of the present manuscript). All other authors and QUEST-EF Investigators have reported that they have no relationships relevant to the content of this paper to disclose.

Figures

Figure 1
Figure 1
Training and testing datasets used for the development and testing of QUEST-EF. A4C, apical four-chamber view; hosp., hospitalization; LVEF, left ventricular ejection fraction; RVEF, right ventricular ejection fraction; WASE, World Alliance of Societies of Echocardiography.
Figure 2
Figure 2
Performance of QUEST-EF. The internal test set for predicting LVEF included 822 A4C videos [LVEF < 50%: 322 (39.2%) videos] from 225 studies [LVEF < 50%: 85 (37.8%) studies], whereas the test set for predicting RVEF comprised 825 A4C videos [RVEF < 45%: 196 (23.8%) videos] from 219 studies [RVEF < 45%: 52 (23.7%) studies]. The external test set for predicting LVEF included 5293 A4C videos [LVEF < 50%: 212 (4.0%) videos] of 1412 patients [LVEF < 50%: 174 (12.3%) patients], whereas the test set for predicting RVEF comprised 5130 A4C videos [RVEF < 45%: 210 (4.1%) videos] of 1249 patients [RVEF < 45%: 87 (7.0%) patients]. The performance of QUEST-EF is reported at the study level (i.e. by averaging the predictions of all A4C videos acquired during the same echocardiographic study). 95% confidence intervals and bands were calculated from 10 000 stratified bootstrap resamples. 3DE, three-dimensional echocardiography; A4C, apical four-chamber view; AUC, area under the receiver operating characteristic curve; LVEF, left ventricular ejection fraction; MAE, mean absolute error; pp, percentage point; RVEF, right ventricular ejection fraction.
See this image and copyright information in PMC

References

    1. Medvedofsky D, Maffessanti F, Weinert L, Tehrani DM, Narang A, Addetia K et al. 2D and 3D echocardiography-derived indices of left ventricular function and shape: relationship with mortality. JACC Cardiovasc Imaging 2018;11:1569–79. - PMC - PubMed
    1. Sayour AA, Tokodi M, Celeng C, Takx RAP, Fábián A, Lakatos BK et al. Association of right ventricular functional parameters with adverse cardiopulmonary outcomes: a meta-analysis. J Am Soc Echocardiogr 2023;36:624–33.e8. - PubMed
    1. Corbett L, O'Driscoll P, Paton M, Oxborough D, Surkova E. Role and application of three-dimensional transthoracic echocardiography in the assessment of left and right ventricular volumes and ejection fraction: a UK nationwide survey. Echo Res Pract 2024;11:8. - PMC - PubMed
    1. Soliman-Aboumarie H, Joshi SS, Cameli M, Michalski B, Manka R, Haugaa K et al. EACVI survey on the multi-modality imaging assessment of the right heart. Eur Heart J Cardiovasc Imaging 2022;23:1417–22. - PubMed
    1. Tokodi M, Magyar B, Soos A, Takeuchi M, Tolvaj M, Lakatos BK et al. Deep learning-based prediction of right ventricular ejection fraction using 2D echocardiograms. JACC Cardiovasc Imaging 2023;16:1005–18. - PubMed