Past and present of computer-assisted dermoscopic diagnosis: performance of a conventional image analyser versus a convolutional neural network in a prospective data set of 1,981 skin lesions
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 tr...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
10 June 2020
|
| In: |
European journal of cancer
Year: 2020, Volume: 135, Pages: 39-46 |
| ISSN: | 1879-0852 |
| DOI: | 10.1016/j.ejca.2020.04.043 |
| Online Access: | Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1016/j.ejca.2020.04.043 Verlag, lizenzpflichtig, Volltext: http://www.sciencedirect.com/science/article/pii/S0959804920302483 
|
| Author Notes: | Katharina Sies, Julia K. Winkler, Christine Fink, Felicitas Bardehle, Ferdinand Toberer, Timo Buhl, Alexander Enk, Andreas Blum, Albert Rosenberger, Holger A. Haenssle |
| Summary: | 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. |
|---|---|
| Item Description: | Gesehen am 27.08.2020 |
| Physical Description: | Online Resource |
| ISSN: | 1879-0852 |
| DOI: | 10.1016/j.ejca.2020.04.043 |