Five critical quality criteria for artificial intelligence-based prediction models

Florien S van Royen; Folkert W Asselbergs; Fernando Alfonso; Panos Vardas; Maarten van Smeden

doi:10.1093/eurheartj/ehad727

Five critical quality criteria for artificial intelligence-based prediction models

Eur Heart J. 2023 Dec 7;44(46):4831-4834. doi: 10.1093/eurheartj/ehad727.

Authors

Florien S van Royen¹, Folkert W Asselbergs^{2

3}, Fernando Alfonso⁴, Panos Vardas⁵, Maarten van Smeden^{6

7}

Affiliations

¹ Department of General Practice & Nursing Science, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.
² Department of Cardiology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, The Netherlands.
³ Health Data Research UK and Institute of Health Informatics, University College London, London, UK.
⁴ Department of Cardiology, Hospital Universitario de la Princesa, Universidad Autónoma de Madrid, IIS-IP. CIVER-CV, Madrid, Spain.
⁵ Biomedical Research Foundation Academy of Athens (BRFAA) and Hygeia Hospitals Group, Athens, Greece.
⁶ Department of Epidemiology & Health Economics, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Universiteitsweg 100, 3584 CG Utrecht, Netherlands.
⁷ Department of Data Science & Biostatistics, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Universiteitsweg 100, 3584 CG Utrecht, The Netherlands.

Abstract

To raise the quality of clinical artificial intelligence (AI) prediction modelling studies in the cardiovascular health domain and thereby improve their impact and relevancy, the editors for digital health, innovation, and quality standards of the European Heart Journal propose five minimal quality criteria for AI-based prediction model development and validation studies: complete reporting, carefully defined intended use of the model, rigorous validation, large enough sample size, and openness of code and software.

Keywords: Artificial intelligence; Diagnosis; Digital health; Prediction; Prognosis.

MeSH terms

Artificial Intelligence*
Heart
Humans
Software*