Past and present of computer-assisted dermoscopic diagnosis

• Verfasst von: Kommoss, Katharina [VerfasserIn]
• Verfasst von: Winkler, Julia K. [VerfasserIn]
• Verfasst von: Müller-Christmann, Christine [VerfasserIn]
• Verfasst von: Niedermair, Felicitas [VerfasserIn]
• Verfasst von: Toberer, Ferdinand [VerfasserIn]
• Verfasst von: Buhl, Timo [VerfasserIn]
• Verfasst von: Enk, Alexander [VerfasserIn]
• Verfasst von: Blum, Andreas [VerfasserIn]
• Verfasst von: Rosenberger, Albert [VerfasserIn]
• Verfasst von: Hänßle, Holger [VerfasserIn]
• Titel: Past and present of computer-assisted dermoscopic diagnosis
• Titelzusatz: performance of a conventional image analyser versus a convolutional neural network in a prospective data set of 1,981 skin lesions
• Verf.angabe: Katharina Sies, Julia K. Winkler, Christine Fink, Felicitas Bardehle, Ferdinand Toberer, Timo Buhl, Alexander Enk, Andreas Blum, Albert Rosenberger, Holger A. Haenssle
• E-Jahr: 2020
• Jahr: 10 June 2020
• Umfang: 8 S.
• Fussnoten: Gesehen am 27.08.2020
• Titel Quelle: Enthalten in: European journal of cancer
• Ort Quelle: Amsterdam [u.a.] : Elsevier, 1992
• Jahr Quelle: 2020
• Band/Heft Quelle: 135(2020), Seite 39-46
• ISSN Quelle: 1879-0852
• DOI: doi:10.1016/j.ejca.2020.04.043
• Datenträger: Online-Ressource
• Sprache: eng
• Sach-SW: Automated melanoma detection
• Sach-SW: Computer-assisted diagnosis
• Sach-SW: Convolutional neural network
• Sach-SW: Deep learning
• Sach-SW: Dermoscopy
• Sach-SW: Skin cancer
• Sach-SW: Skin lesions
• K10plus-PPN: 172791290X

Abstract

Background - Convolutional neural networks (CNNs) have shown a dermatologist-level performance in the classification of skin lesions. We aimed to deliver a head-to-head comparison of a conventional image analyser (CIA), which depends on segmentation and weighting of handcrafted features, to a CNN trained by deep learning. - Methods - Cross-sectional study using a real-world, prospectively acquired, dermoscopic dataset of 1981 skin lesions to compare the diagnostic performance of a market-approved CNN (Moleanalyzer-Pro™, developed in 2018) to a CIA (Moleanalyzer-3™/Dynamole™; developed in 2004, all FotoFinder Systems Inc, Germany). As a reference standard, we used histopathological diagnoses (n = 785) or, in non-excised benign lesions (n = 1196), expert consensus plus an uneventful follow-up by sequential digital dermoscopy for at least 2 years. - Results - A total of 281 malignant lesions and 1700 benign lesions from 435 patients (62.2% male, mean age: 52 years) were prospectively imaged. The CNN showed a sensitivity of 77.6% (95% confidence interval [CI]: [72.4%-82.1%]), specificity of 95.3% (95% CI: [94.2%-96.2%]), and receiver operating characteristic (ROC)-area under the curve (AUC) of 0.945 (95% CI: [0.930-0.961]). In contrast, the CIA achieved a sensitivity of 53.4% (95% CI: [47.5%-59.1%]), specificity of 86.6% (95% CI: [84.9%-88.1%]) and ROC-AUC of 0.738 (95% CI: [0.701-0.774]). The data set included melanomas originally diagnosed by dynamic changes during sequential digital dermoscopy (52 of 201, 20.6%), which reduced the sensitivities of both classifiers. Pairwise comparisons of sensitivities, specificities, and ROC-AUCs indicated a clear outperformance by the CNN (all p < 0.001). - Conclusions - The superior diagnostic performance of the CNN argues against a continued application of former CIAs as an aide to physicians’ clinical management decisions.