Skip to content
DraconiticResearch
Search Sources API
203,203 studies31 sourcesUpdated 31 min ago
Peer-reviewedCC BY-NCEuropean heart journal. Digital health

Deep learning-based echocardiography for cardiac assessment: a systematic literature review and meta-analysis.

  • Aida Romano-Martinez1,2,3
  • Silav Zeid1,3,4
  • Mykhailo Krolevets1,2,3
  • Konstantin Kontohow-Beckers1
  • Vincent Ten Cate1,3,4
  • Philipp Lurz3,5
  • Sandy Engelhardt6,7
  • Stefan Kramer8
  • Sven-Oliver Troebs1,9
  • Philipp S Wild1,2,3,4
9 affiliations
  1. 1Preventive Cardiology and Preventive Medicine, Department of Cardiology, University Medical Centre of the Johannes Gutenberg-University, Mainz Langenbeckstr. 1, Mainz 55131, Germany.
  2. 2Systems Medicine, Institute of Molecular Biology (IMB), Mainz, Germany.
  3. 3German Centre for Cardiovascular Research (DZHK), Partner Site Rhine-Main, Mainz, Germany.
  4. 4Clinical Epidemiology and Systems Medicine, Centre for Thrombosis and Hemostasis (CTH), University Medical Centre of the Johannes Gutenberg-University, Mainz, Germany.
  5. 5Centre for Cardiology-Cardiology I, University Medical Centre of the Johannes Gutenberg-University, Mainz, Germany.
  6. 6Department of Cardiology, Angiology and Pneumology, Heidelberg University Hospital, Heidelberg, Germany.
  7. 7German Centre for Cardiovascular Research (DZHK), Partner Site Heidelberg/Mannheim, Heidelberg, Germany.
  8. 8Institute of Computer Science, Johannes Gutenberg University, Mainz, Germany.
  9. 9Boehringer Ingelheim International GmbH, Ingelheim am Rhein, Germany.
Read full textPDFDOICite

Abstract

Aims

Artificial intelligence, particularly deep learning (DL), has shown significant potential for enhancing the assessment of echocardiograms in clinical practice. This systematic review and meta-analysis aimed to evaluate the performance of DL models and to identify factors affecting their performance.Methods and results

A systematic review of the PubMed, CENTRAL, and Web of Science databases was conducted, yielding 16 248 potential publications. Of these, 160 studies met the inclusion criteria. Only studies applying fully automated DL networks to echocardiographic loops were considered. Performance estimates were pooled using a random-effects meta-analysis. The studies were categorized into five distinct groups: prediction of disease or mortality, replication of established measures, segmentation of cardiac chambers, classification of chamber views, and assessment of image quality. High heterogeneity that could not be resolved was observed in studies focusing on cardiac chamber segmentation and classification. Model performance was influenced by a variety of different factors, including medical target and input characteristics such as video resolution, number of frames used, and total participants included during training.Conclusion

DL models achieve high performance across various applications in echocardiography. In selected tasks, they may reduce reliance on expert input and support greater automation of echocardiographic workflows. However, further validation, assessment of clinical utility, and evaluation in real-world settings are required before widespread clinical adoption.

Related

  1. Transthoracic echocardiography vs. cardiac magnetic resonance imaging for the diagnosis of left ventricular thrombus: an updated systematic review and diagnostic accuracy meta-analysis
    European heart journal. Imaging methods and practicePeer-reviewedCC BYFull textGeneral medicineEurope PMC
  2. Efficacy and safety of mechanical chest compression compared with manual cardiopulmonary resuscitation in adult non-traumatic cardiac arrest: a systematic review of the literature
    Sara Rizzo, Domitilla Nistri, Mauro Carmine Martucci +3
    Dissertation NursingPeer-reviewedCC BY-NC-NDFull textGeneral medicineDOAJ
  3. Advancements in echocardiography for congenital heart defects (CHDs): a literature review
    Muhammad Abdullah Humayun, Carla Jammal, Anita Zhou +6
    Annals of medicine and surgery (2012)Peer-reviewedCC BY-NC-NDFull textGeneral medicineEurope PMC
  4. Complex cardiac schwannoma surgery: Navigation with perioperative 3-dimensional transesophageal echocardiography
    Jayakumar Thanathu Krishnan Nair, Manjusha N. Pillai, Adinarayanan Kishore Nochur +1
    JTCVS techniquesPeer-reviewedCC BY-NC-NDFull textGeneral medicineEurope PMC
  5. Cardiac Capillary Hemangioma in a Patient Presenting with Syncope: Report of a Case and Systematic Review of the Literature
    George Bahna, Abbas Agaimy, Robert Cesnjevar +6
    Journal of Clinical MedicinePeer-reviewedCC BY 4.0Cardiovascular medicineCrossref

Cite

BibTeXRIS
BibTeX
@article{romanomartinez2026deep,
  title = {{Deep learning-based echocardiography for cardiac assessment: a systematic literature review and meta-analysis.}},
  author = {Aida Romano-Martinez and Silav Zeid and Mykhailo Krolevets and Konstantin Kontohow-Beckers and Vincent Ten Cate and Philipp Lurz and Sandy Engelhardt and Stefan Kramer and Sven-Oliver Troebs and Philipp S Wild},
  journal = {European heart journal. Digital health},
  year = {2026},
  month = oct,
  doi = {10.1093/ehjdh/ztag148},
  url = {https://europepmc.org/article/MED/42899077},
  note = {CC BY-NC},
}
RIS
TY  - JOUR
TI  - Deep learning-based echocardiography for cardiac assessment: a systematic literature review and meta-analysis.
AU  - Aida Romano-Martinez
AU  - Silav Zeid
AU  - Mykhailo Krolevets
AU  - Konstantin Kontohow-Beckers
AU  - Vincent Ten Cate
AU  - Philipp Lurz
AU  - Sandy Engelhardt
AU  - Stefan Kramer
AU  - Sven-Oliver Troebs
AU  - Philipp S Wild
PY  - 2026
DA  - 2026/10/19
JO  - European heart journal. Digital health
DO  - 10.1093/ehjdh/ztag148
UR  - https://europepmc.org/article/MED/42899077
AB  - Aims  Artificial intelligence, particularly deep learning (DL), has shown significant potential for enhancing the assessment of echocardiograms in clinical practice. This systematic review and meta-analysis aimed to evaluate the performance of DL models and to identify factors affecting their performance.Methods and results  A systematic review of the PubMed, CENTRAL, and Web of Science databases was conducted, yielding 16 248 potential publications. Of these, 160 studies met the inclusion criteria. Only studies applying fully automated DL networks to echocardiographic loops were considered. Performance estimates were pooled using a random-effects meta-analysis. The studies were categorized into five distinct groups: prediction of disease or mortality, replication of established measures, segmentation of cardiac chambers, classification of chamber views, and assessment of image quality. High heterogeneity that could not be resolved was observed in studies focusing on cardiac chamber segmentation and classification. Model performance was influenced by a variety of different factors, including medical target and input characteristics such as video resolution, number of frames used, and total participants included during training.Conclusion  DL models achieve high performance across various applications in echocardiography. In selected tasks, they may reduce reliance on expert input and support greater automation of echocardiographic workflows. However, further validation, assessment of clinical utility, and evaluation in real-world settings are required before widespread clinical adoption.
SN  - 2634-3916
ER  -