Diagnosis melanoma with artificial intelligence systems: A meta‐analysis study and systematic review
Diagnose von Melanom mit künstlichen Intelligenzsystemen: Meta-Analyse und systematische Übersicht
Journal of the European Academy of Dermatology and Venereology · 2025-11 · 39 (11) · 1912-1922
DOI: 10.1111/jdv.20781 ↗Auf einen Blick
39 Primärstudien zu KI-gestützten Verfahren für die Melanomdiagnose zeigten eine hohe diagnostische Leistung. Das gepoolte AUC lag bei 0,96; Hybrid-Modelle erreichten 0,98.
Fragestellung
Wie gut erkennen KI- und Machine-Learning-Algorithmen Melanome auf dermoskopischen Bildern?
Methode
Systematische Übersicht und Metaanalyse nach PRISMA; Suche in Web of Science, Scopus und MEDLINE von Januar 2006 bis Mai 2024.
Ergebnisse
39 Primärstudien mit 78 diagnostischen Tests. Gepooltes AUC 0,96 (95 %-KI 0,94–0,98), Sensitivität 0,89 (95 %-KI 0,85–0,91), Spezifität 0,92 (95 %-KI 0,90–0,94). Hybrid-Modelle: AUC 0,98 (95 %-KI 0,96–0,99).
Schlussfolgerung der Autor:innen
Die Autor:innen sehen KI- und Machine-Learning-Algorithmen als nützlich zur Unterstützung der Melanomdiagnose; Hybrid-Modelle zeigten eine hohe Leistung.
Kernaussagen
- AUC gesamt: 0,96 (95 %-KI 0,94–0,98)
- Sensitivität 0,89 und Spezifität 0,92
- Hybrid-Modelle: AUC 0,98 (95 %-KI 0,96–0,99)
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Systematische Übersicht in der Zeitschrift Journal of the European Academy of Dermatology and Venereology, online erschienen am 06.06.2025. Verfasst von Gözde Ertürk Zararsız und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biostatistics, Erciyes University School of Medicine, Kayseri, Turkey. Hauptthemen laut MEDLINE: Melanoma, Artificial Intelligence, Skin Neoplasms. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Erciyes University Scientific Research Coordination Unit.
Abstract (Originalsprache)
Background
One of the most promising and rapidly advancing research areas in recent years is using dermoscopic images for automatic diagnosis with artificial intelligence and machine learning methods.
Objectives
This study aimed to synthesize the existing studies for the clinical use of applications made with artificial intelligence methods and to summarize the predictive performance of deep learning and hybrid models‐based algorithms in all these studies with a large‐scale meta‐analysis.
Method
The literature review was conducted between January 2006 and May 2024, and meta‐analysis data were created by scanning the Web of Science (WOS), Scopus and MEDLINE databases. This study followed the Preferred Reporting Items for Systematic Reviews and Meta‐Analysis (PRISMA) checklist.
Results
A total of 2722 articles were evaluated. Data from 78 diagnostic tests from 39 primary studies meeting the inclusion and exclusion criteria were assessed. The pooled SROC overall model AUC was 0.96 [95% CI: 0.94–0.98], sensitivity was 0.89 [95% CI: 0.85–0.91] and specificity was 0.92 [95% CI: 0.90–0.94]. In the subgroup analyses, the pooled AUC was 0.98 [95% CI: 0.96–0.99] for HYBRID models.
Conclusions
Recent studies have suggested that artificial intelligence algorithms and machine learning methods should be used extensively in medicine to assist physicians, especially in diagnosing melanoma. The ability of HYBRID model algorithms to predict diseases is promising. In particular, the performance of HYBRID models was found to be high. This information can assist clinicians in interpreting the most appropriate algorithms for diagnosing melanoma.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Biostatistics, Erciyes University School of Medicine, Kayseri, Turkey.
- Department of Biometrics, Erciyes University, Faculty of Veterinary Medicine, Kayseri, Turkey.
- Department of Economics and Administrative Programs, Hacettepe University Vocational School of Social Sciences, Ankara, Turkey.
- Department of Dermatology, Faculty of Medicine, Kocaeli University, Kocaeli, Turkey.
- Chronic Skin Diseases Unit, Department of Dermatology, University of Health Sciences, Kayseri City Education and Research Hospital, Kayseri, Turkey.
- Department of Molecular Biology and Genetics, Faculty of Life and Natural Sciences, Abdullah Gul University, Kayseri, Turkey.
ORCID-Profile: Gözde Ertürk Zararsız, Serra Ilayda Yerlitaş Taştan, Abdullah Demirbaş, Ömer Faruk Elmas
Förderung
- Erciyes University Scientific Research Coordination Unit
Zitieren
APA
Ertürk Zararsız, G., Yerlitaş Taştan, S. I., Çelik Gürbulak, E., Erakcaoğlu, A., Yılmaz Işıkhan, S., Demirbaş, A., Ertaş, R., Eroğlu, İ., Korkmaz, S., Elmas, Ö. F., & Zararsız, G. (2025). Diagnosis melanoma with artificial intelligence systems: A meta‐analysis study and systematic review. Journal of the European Academy of Dermatology and Venereology, 39(11), 1912-1922. https://doi.org/10.1111/jdv.20781
Vancouver
Ertürk Zararsız G, Yerlitaş Taştan SI, Çelik Gürbulak E, Erakcaoğlu A, Yılmaz Işıkhan S, Demirbaş A, et al. Diagnosis melanoma with artificial intelligence systems: A meta‐analysis study and systematic review. Journal of the European Academy of Dermatology and Venereology. 2025;39(11):1912-1922. doi:10.1111/jdv.20781
BibTeX
@article{ErtrkZararsz2025,
author = {Ertürk Zararsız, Gözde and Yerlitaş Taştan, Serra Ilayda and Çelik Gürbulak, Elif and Erakcaoğlu, Aleyna and Yılmaz Işıkhan, Selen and Demirbaş, Abdullah and Ertaş, Ragıp and Eroğlu, İrem and Korkmaz, Selçuk and Elmas, Ömer Faruk and Zararsız, Gökmen},
title = {Diagnosis melanoma with artificial intelligence systems: A meta‐analysis study and systematic review},
journal = {Journal of the European Academy of Dermatology and Venereology},
year = {2025},
volume = {39},
number = {11},
pages = {1912--1922},
doi = {10.1111/jdv.20781}
}
EUVIMED