Artificial Intelligence Applied to Non-Invasive Imaging Modalities in Identification of Nonmelanoma Skin Cancer: A Systematic Review
Künstliche Intelligenz in nicht-invasiven Bildgebungsverfahren zur Identifikation von Nichtmelanom-Hautkrebs: systematische Übersicht
Cancers · 01.02.2024 · 16 (3) · 629
DOI: 10.3390/cancers16030629 ↗Auf einen Blick
44 Studien zeigen eine insgesamt günstige Leistung von KI-Algorithmen in der Diagnose von Nichtmelanom-Hautkrebs über nicht-invasive Bildgebung. Die mittlere Genauigkeit lag bei 86,80 %.
Fragestellung
Wie gut sind KI-Modelle in nicht-invasiven Bildgebungsverfahren zur frühen Erkennung von Nichtmelanom-Hautkrebs?
Methode
Systematische Übersicht von peer-reviewten Studien in MEDLINE, Cochrane, PubMed und Embase; Studien 2018 bis 2023; Biasbewertung mit QUADAS-2.
Ergebnisse
44 Studien: 40 zu Dermatoskopie, 3 zu RCM, 1 zu HEI. Mittlere Genauigkeit aller Modalitäten 86,80 %; bei Dermatoskopie ebenfalls 86,80 %; bei RCM 87 % in 1 Studie.
Einschränkungen
Weitere Forschung nötig; viele Testdatensätze enthielten auch Melanom und andere pigmentierte Läsionen.
Schlussfolgerung der Autor:innen
KI-Algorithmen zeigten insgesamt eine günstige Leistung bei der Diagnose von Nichtmelanom-Hautkrebs über nicht-invasive Bildgebung; weitere Forschung ist nötig.
Kernaussagen
- Mittlere Genauigkeit aller Modalitäten: 86,80 %
- 44 Studien eingeschlossen
- Testdatensätze enthielten oft auch Melanom und andere Läsionen
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Übersichtsarbeit in der Zeitschrift Cancers, online erschienen am 01.02.2024. Verfasst von Emilie A Foltz und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Dermatology, Oregon Health & Science University, Portland, OR 97201, USA. Die Arbeit wurde bisher 31-mal zitiert (Crossref) und nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background
The objective of this study is to systematically analyze the current state of the literature regarding novel artificial intelligence (AI) machine learning models utilized in non-invasive imaging for the early detection of nonmelanoma skin cancers. Furthermore, we aimed to assess their potential clinical relevance by evaluating the accuracy, sensitivity, and specificity of each algorithm and assessing for the risk of bias.
Methods
Two reviewers screened the MEDLINE, Cochrane, PubMed, and Embase databases for peer-reviewed studies that focused on AI-based skin cancer classification involving nonmelanoma skin cancers and were published between 2018 and 2023. The search terms included skin neoplasms, nonmelanoma, basal-cell carcinoma, squamous-cell carcinoma, diagnostic techniques and procedures, artificial intelligence, algorithms, computer systems, dermoscopy, reflectance confocal microscopy, and optical coherence tomography. Based on the search results, only studies that directly answered the review objectives were included and the efficacy measures for each were recorded. A QUADAS-2 risk assessment for bias in included studies was then conducted.
Results
A total of 44 studies were included in our review; 40 utilizing dermoscopy, 3 using reflectance confocal microscopy (RCM), and 1 for hyperspectral epidermal imaging (HEI). The average accuracy of AI algorithms applied to all imaging modalities combined was 86.80%, with the same average for dermoscopy. Only one of the three studies applying AI to RCM measured accuracy, with a result of 87%. Accuracy was not measured in regard to AI based HEI interpretation.
Conclusion
AI algorithms exhibited an overall favorable performance in the diagnosis of nonmelanoma skin cancer via noninvasive imaging techniques. Ultimately, further research is needed to isolate pooled diagnostic accuracy for nonmelanoma skin cancers as many testing datasets also include melanoma and other pigmented lesions.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Dermatology, Oregon Health & Science University, Portland, OR 97201, USA.
- Elson S. Floyd College of Medicine, Washington State University, Spokane, WA 99202, USA.
- John A. Burns School of Medicine, University of Hawai'i at Manoa, Honolulu, HI 96813, USA.
- Biostatistics Shared Resource, Knight Cancer Institute, Oregon Health & Science University, Portland, OR 97201, USA.
ORCID-Profile: Emilie A Foltz, Joanna Ludzik
Zitieren
APA
Foltz, E. A., Witkowski, A., Becker, A. L., Latour, E., Lim, J. Y., Hamilton, A., & Ludzik, J. (2024). Artificial Intelligence Applied to Non-Invasive Imaging Modalities in Identification of Nonmelanoma Skin Cancer: A Systematic Review. Cancers, 16(3), 629. https://doi.org/10.3390/cancers16030629
Vancouver
Foltz EA, Witkowski A, Becker AL, Latour E, Lim JY, Hamilton A, et al. Artificial Intelligence Applied to Non-Invasive Imaging Modalities in Identification of Nonmelanoma Skin Cancer: A Systematic Review. Cancers. 2024;16(3):629. doi:10.3390/cancers16030629
BibTeX
@article{Foltz2024,
author = {Foltz, Emilie A and Witkowski, Alexander and Becker, Alyssa L and Latour, Emile and Lim, Jeong Youn and Hamilton, Andrew and Ludzik, Joanna},
title = {Artificial Intelligence Applied to Non-Invasive Imaging Modalities in Identification of Nonmelanoma Skin Cancer: A Systematic Review},
journal = {Cancers},
year = {2024},
volume = {16},
number = {3},
pages = {629},
doi = {10.3390/cancers16030629}
}
EUVIMED