Artificial intelligence in cardiology: an evidence map across subspecialties, 2014-26.
28 affiliations
- 11st Department of Cardiology, Medical University of Warsaw, Banacha 1A, 02-097 Warsaw, Poland.
- 2Computational Physics Laboratory, Tampere University, Tampere, Finland.
- 3Department of Clinical Sciences, Danderyd Hospital, Karolinska Institutet, Stockholm, Sweden.
- 4Cardiology Department, University Hospital Virgen de la Victoria, Instituto de Investigación Biomédica de Málaga (IBIMA), Málaga, Spain.
- 5Centro de Investigación Biomédica en Red en Enfermedades Cardiovasculares (CIBERCV), Instituto de Salud Carlos III, Madrid, Spain.
- 6Department of Biomedical and Dental Sciences and of Morphological and Functional Images, University of Messina, Messina, Italy.
- 7Health Sciences University Faculty of Medicine, Department of Cardiology, Istanbul, Türkiye.
- 8University of Crete, School of Medicine, Heraklion, Crete, Greece.
- 9Cardiac Imaging Unit, Complexo Hospitalario Universitario de A Coruña (CHUAC), A Coruña, Spain.
- 10Thorax Center, Department of Cardiology, Erasmus MC, Cardiovascular Institute, Rotterdam, The Netherlands.
- 11Department of Medicine, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Sweden.
- 12Digital Medicine & Robotics Center, Jagiellonian University Medical College, Krakow, Poland.
- 13Cardiology Department, Centrum Voor Hart en Vaatziekten (CHVZ), Universitair Ziekenhuis Brussel, Brussels, Belgium.
- 14Department of Cardiology, H.U.B.-Hôpital Erasme, Université Libre de Bruxelles, Brussels, Belgium.
- 15Department of Cardiology, Máxima Medical Centre, Veldhoven, The Netherlands.
- 16Department of Industrial Design, Eindhoven University of Technology, Eindhoven, The Netherlands.
- 17Cardio Pulmonary Department, IRCCS San Raffaele Roma, Rome, Italy.
- 18Department of Human Science and Promotion of Quality of Life, San Raffaele Open University, Rome, Italy.
- 19Clinical Cardiology Center, National Institute of Medicine MSWiA, Warsaw, Poland.
- 20Center for Postgraduate Medical Education, Warsaw, Poland.
- 21Department of Cardiology, Antwerp University Hospital, Antwerp, Belgium.
- 22Research Group Cardiovascular Diseases and Center for Research and Innovation in Care, University of Antwerp, Antwerp, Belgium.
- 23Faculty of Medicine and Life Sciences, Hasselt University, Hasselt, Belgium.
- 24Department V Internal Medicine, University of Medicine and Pharmacy Victor Babeș Timișoara, Research Centre, Institute of Cardiovascular Diseases Timișoara, Romania.
- 25German Heart Center Charité, Cardiology Charité Mitte, Berlin, Germany.
- 26Department of Cardiology, Amsterdam Cardiovascular Sciences, Amsterdam University Medical Centre, University of Amsterdam, Amsterdam, The Netherlands.
- 27National Institute for Health Research University College London Hospitals Biomedical Research Centre, University College London, London, UK.
- 28Institute of Health Informatics, University College London, London, UK.
Abstract
Aims
Artificial intelligence (AI) is increasingly being applied in healthcare, including cardiology, with growing interest in its potential to support diagnosis, prognosis, and clinical decision-making. The rapid pace of technological advancements is not matched by equivalent progress in the regulation and appraisal of these systems. This review aims to provide a structured evidence map of AI models in cardiology, by grading the robustness of clinical validation and mapping it against the quantity of publications across main subspecialties.Methods and results
To assess the strength of evidence, we defined, developed, and applied a novel classification tailored for AI models that ranks evidence into five Grades of Validation according to the level of clinical validity. We screened literature from 2014 to 2026 according to the PRISMA guidelines, identifying a total of 2807 articles on AI applications within cardiology, 490 of which met the inclusion criteria (validation was performed in at least one external, independent patient population), revealing a highly heterogeneous distribution of evidence across subspecialties. Most studies in AI focused on improved diagnostics (n = 322, 65.7%); however, advanced integration in prognostic (n = 122, 24.9%) and therapeutic applications remains limited (n = 46, 9.4%). Cardiovascular imaging leads AI applications, especially in coronary computed tomography and echocardiography, followed by non-invasive electrophysiology, predominantly in atrial fibrillation screening and ventricular dysfunction detection.Conclusion
The scarcity of prospective validations across different resources and geographical settings (n = 134, 27.3%) and randomized controlled trials (n = 35, 3.8%) limits the adoption by clinical guidelines and widespread implementation of AI algorithms.
Related
- Artificial intelligence across the gastric cancer care continuum: Clinical evidence, multimodal integration, and translation
- Psychometric characterization of human and artificial intelligence performance on cardiology residency in-service examination items
- Female leadership across ophthalmology subspecialties: a cross-sectional study of academic leadership roles
- Low-Dose CT in the Age of Artificial Intelligence: Techniques, Evidence, and Clinical Impact
- Bias Mitigation Across the Healthcare Artificial Intelligence Lifecycle: A Structured Narrative Review
BibTeX
@article{koltowski2026artificial,
title = {{Artificial intelligence in cardiology: an evidence map across subspecialties, 2014-26.}},
author = {Łukasz Kołtowski and Mikołaj Basza and Mateusz Soliński and Bartosz Rolek and Helge Brandberg and Francesco Costa and Nurgul Keser and Maria Marketou and Marcin Grabowski and Rafael Vidal-Perez and Nico Bruining and Emma Svennberg and Dariusz Dudek and Bernard Cosyns and Paweł Balsam and Ruben Casado-Arroyo and Hareld Kemps and Maurizio Volterrani and Piotr Szymański and Lien Desteghe and Ruxandra Christodorescu and Grzegorz Opolski and Gerhard Hindricks and Folkert W Asselbergs},
journal = {European heart journal. Digital health},
year = {2026},
month = sep,
doi = {10.1093/ehjdh/ztag152},
url = {https://europepmc.org/article/MED/42909890},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - Artificial intelligence in cardiology: an evidence map across subspecialties, 2014-26.
AU - Łukasz Kołtowski
AU - Mikołaj Basza
AU - Mateusz Soliński
AU - Bartosz Rolek
AU - Helge Brandberg
AU - Francesco Costa
AU - Nurgul Keser
AU - Maria Marketou
AU - Marcin Grabowski
AU - Rafael Vidal-Perez
AU - Nico Bruining
AU - Emma Svennberg
AU - Dariusz Dudek
AU - Bernard Cosyns
AU - Paweł Balsam
AU - Ruben Casado-Arroyo
AU - Hareld Kemps
AU - Maurizio Volterrani
AU - Piotr Szymański
AU - Lien Desteghe
AU - Ruxandra Christodorescu
AU - Grzegorz Opolski
AU - Gerhard Hindricks
AU - Folkert W Asselbergs
PY - 2026
DA - 2026/09/23
JO - European heart journal. Digital health
DO - 10.1093/ehjdh/ztag152
UR - https://europepmc.org/article/MED/42909890
AB - Aims Artificial intelligence (AI) is increasingly being applied in healthcare, including cardiology, with growing interest in its potential to support diagnosis, prognosis, and clinical decision-making. The rapid pace of technological advancements is not matched by equivalent progress in the regulation and appraisal of these systems. This review aims to provide a structured evidence map of AI models in cardiology, by grading the robustness of clinical validation and mapping it against the quantity of publications across main subspecialties.Methods and results To assess the strength of evidence, we defined, developed, and applied a novel classification tailored for AI models that ranks evidence into five Grades of Validation according to the level of clinical validity. We screened literature from 2014 to 2026 according to the PRISMA guidelines, identifying a total of 2807 articles on AI applications within cardiology, 490 of which met the inclusion criteria (validation was performed in at least one external, independent patient population), revealing a highly heterogeneous distribution of evidence across subspecialties. Most studies in AI focused on improved diagnostics (n = 322, 65.7%); however, advanced integration in prognostic (n = 122, 24.9%) and therapeutic applications remains limited (n = 46, 9.4%). Cardiovascular imaging leads AI applications, especially in coronary computed tomography and echocardiography, followed by non-invasive electrophysiology, predominantly in atrial fibrillation screening and ventricular dysfunction detection.Conclusion The scarcity of prospective validations across different resources and geographical settings (n = 134, 27.3%) and randomized controlled trials (n = 35, 3.8%) limits the adoption by clinical guidelines and widespread implementation of AI algorithms.
SN - 2634-3916
ER -