Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations

Robust assessment of artificial intelligence (AI) models in medical imaging is paramount for reliable clinical integration. This international collaborative review paper provides an overview of key evaluation metrics across diverse tasks, including classification, regression, survival analysis, dete...

Descripción completa

Detalles Bibliográficos
Autores: Kocak, Burak, Klontzas, Michail E., Stanzione, Arnaldo, Meddeb, Aymen, Demircioğlu, Aydın, Bluethgen, Christian, Bressem, Keno K., Ugga, Lorenzo, Mercaldo, Nathaniel, Díaz, Oliver, Cuocolo, Renato
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2025
País:España
Institución:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
Repositorio:Recercat. Dipósit de la Recerca de Catalunya
OAI Identifier:oai:recercat.cat:2445/227851
Acceso en línea:https://hdl.handle.net/2445/227851
Access Level:acceso abierto
Palabra clave:Intel·ligència artificial en medicina
Diagnòstic per la imatge
Aprenentatge automàtic
Algorismes computacionals
Medical artificial intelligence
Diagnostic imaging
Machine learning
Computer algorithms
id ES_e71f723e95a9072ddd1e5227acb2ccbd
oai_identifier_str oai:recercat.cat:2445/227851
network_acronym_str ES
network_name_str España
repository_id_str
spelling Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendationsKocak, BurakKlontzas, Michail E.Stanzione, ArnaldoMeddeb, AymenDemircioğlu, AydınBluethgen, ChristianBressem, Keno K.Ugga, LorenzoMercaldo, NathanielDíaz, OliverCuocolo, RenatoIntel·ligència artificial en medicinaDiagnòstic per la imatgeAprenentatge automàticAlgorismes computacionalsMedical artificial intelligenceDiagnostic imagingMachine learningComputer algorithmsRobust assessment of artificial intelligence (AI) models in medical imaging is paramount for reliable clinical integration. This international collaborative review paper provides an overview of key evaluation metrics across diverse tasks, including classification, regression, survival analysis, detection, and segmentation, as well as specialized metrics for calibration, foundation models, large language models, and synthetic images. Challenges of comparing models statistically and translating metric scores to clinical practice are also discussed. For each section, the paper outlines fundamental metrics, identifies common pitfalls and misapplications, and offers recommendations for more robust evaluations. Key recommendations often involve utilizing multiple, complementary metrics tailored to the specific task and dataset properties, transparent reporting of methodology, and critically, considering the clinical utility and real-world implications of model performance. Ultimately, effective evaluation requires a comprehensive, context-aware approach that goes beyond statistical metrics to ensure.Elsevier B.V.2026202620252026info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersion24 p.application/pdfhttps://hdl.handle.net/2445/227851Articles publicats en revistes (Matemàtiques i Informàtica)reponame:Recercat. Dipósit de la Recerca de Catalunyainstname:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)InglésReproducció del document publicat a: https://doi.org/10.1016/j.ejrai.2025.100030European Journal of Radiology Artificial Intelligence, 2025, vol. 3, p. 100030https://doi.org/10.1016/j.ejrai.2025.100030cc-by (c) Burak Kocak et al., 2025http://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/openAccessoai:recercat.cat:2445/2278512026-05-29T05:05:01Z
dc.title.none.fl_str_mv Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations
title Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations
spellingShingle Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations
Kocak, Burak
Intel·ligència artificial en medicina
Diagnòstic per la imatge
Aprenentatge automàtic
Algorismes computacionals
Medical artificial intelligence
Diagnostic imaging
Machine learning
Computer algorithms
title_short Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations
title_full Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations
title_fullStr Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations
title_full_unstemmed Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations
title_sort Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations
dc.creator.none.fl_str_mv Kocak, Burak
Klontzas, Michail E.
Stanzione, Arnaldo
Meddeb, Aymen
Demircioğlu, Aydın
Bluethgen, Christian
Bressem, Keno K.
Ugga, Lorenzo
Mercaldo, Nathaniel
Díaz, Oliver
Cuocolo, Renato
author Kocak, Burak
author_facet Kocak, Burak
Klontzas, Michail E.
Stanzione, Arnaldo
Meddeb, Aymen
Demircioğlu, Aydın
Bluethgen, Christian
Bressem, Keno K.
Ugga, Lorenzo
Mercaldo, Nathaniel
Díaz, Oliver
Cuocolo, Renato
author_role author
author2 Klontzas, Michail E.
Stanzione, Arnaldo
Meddeb, Aymen
Demircioğlu, Aydın
Bluethgen, Christian
Bressem, Keno K.
Ugga, Lorenzo
Mercaldo, Nathaniel
Díaz, Oliver
Cuocolo, Renato
author2_role author
author
author
author
author
author
author
author
author
author
dc.subject.none.fl_str_mv Intel·ligència artificial en medicina
Diagnòstic per la imatge
Aprenentatge automàtic
Algorismes computacionals
Medical artificial intelligence
Diagnostic imaging
Machine learning
Computer algorithms
topic Intel·ligència artificial en medicina
Diagnòstic per la imatge
Aprenentatge automàtic
Algorismes computacionals
Medical artificial intelligence
Diagnostic imaging
Machine learning
Computer algorithms
description Robust assessment of artificial intelligence (AI) models in medical imaging is paramount for reliable clinical integration. This international collaborative review paper provides an overview of key evaluation metrics across diverse tasks, including classification, regression, survival analysis, detection, and segmentation, as well as specialized metrics for calibration, foundation models, large language models, and synthetic images. Challenges of comparing models statistically and translating metric scores to clinical practice are also discussed. For each section, the paper outlines fundamental metrics, identifies common pitfalls and misapplications, and offers recommendations for more robust evaluations. Key recommendations often involve utilizing multiple, complementary metrics tailored to the specific task and dataset properties, transparent reporting of methodology, and critically, considering the clinical utility and real-world implications of model performance. Ultimately, effective evaluation requires a comprehensive, context-aware approach that goes beyond statistical metrics to ensure.
publishDate 2025
dc.date.none.fl_str_mv 2025
2026
2026
2026
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv https://hdl.handle.net/2445/227851
url https://hdl.handle.net/2445/227851
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Reproducció del document publicat a: https://doi.org/10.1016/j.ejrai.2025.100030
European Journal of Radiology Artificial Intelligence, 2025, vol. 3, p. 100030
https://doi.org/10.1016/j.ejrai.2025.100030
dc.rights.none.fl_str_mv cc-by (c) Burak Kocak et al., 2025
http://creativecommons.org/licenses/by/4.0/
info:eu-repo/semantics/openAccess
rights_invalid_str_mv cc-by (c) Burak Kocak et al., 2025
http://creativecommons.org/licenses/by/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv 24 p.
application/pdf
dc.publisher.none.fl_str_mv Elsevier B.V.
publisher.none.fl_str_mv Elsevier B.V.
dc.source.none.fl_str_mv Articles publicats en revistes (Matemàtiques i Informàtica)
reponame:Recercat. Dipósit de la Recerca de Catalunya
instname:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
instname_str Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
reponame_str Recercat. Dipósit de la Recerca de Catalunya
collection Recercat. Dipósit de la Recerca de Catalunya
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869422826804477952
score 15,812429

Similares en LA Referencia