1. |
- Haenssle, H A, et al.
(author)
-
Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists.
- 2018
-
In: Annals of Oncology. - : Elsevier BV. - 1569-8041 .- 0923-7534. ; 29:8, s. 1836-1842
-
Journal article (peer-reviewed)abstract
- Deep learning convolutional neural networks (CNN) may facilitate melanoma detection, but data comparing a CNN's diagnostic performance to larger groups of dermatologists are lacking.Google's Inception v4 CNN architecture was trained and validated using dermoscopic images and corresponding diagnoses. In a comparative cross-sectional reader study a 100-image test-set was used (level-I: dermoscopy only; level-II: dermoscopy plus clinical information and images). Main outcome measures were sensitivity, specificity and area under the curve (AUC) of receiver operating characteristics (ROC) for diagnostic classification (dichotomous) of lesions by the CNN versus an international group of 58 dermatologists during level-I or -II of the reader study. Secondary end points included the dermatologists' diagnostic performance in their management decisions and differences in the diagnostic performance of dermatologists during level-I and -II of the reader study. Additionally, the CNN's performance was compared with the top-five algorithms of the 2016 International Symposium on Biomedical Imaging (ISBI) challenge.In level-I dermatologists achieved a mean (±standard deviation) sensitivity and specificity for lesion classification of 86.6% (±9.3%) and 71.3% (±11.2%), respectively. More clinical information (level-II) improved the sensitivity to 88.9% (±9.6%, P=0.19) and specificity to 75.7% (±11.7%, P<0.05). The CNN ROC curve revealed a higher specificity of 82.5% when compared with dermatologists in level-I (71.3%, P<0.01) and level-II (75.7%, P<0.01) at their sensitivities of 86.6% and 88.9%, respectively. The CNN ROC AUC was greater than the mean ROC area of dermatologists (0.86 versus 0.79, P<0.01). The CNN scored results close to the top three algorithms of the ISBI 2016 challenge.For the first time we compared a CNN's diagnostic performance with a large international group of 58 dermatologists, including 30 experts. Most dermatologists were outperformed by the CNN. Irrespective of any physicians' experience, they may benefit from assistance by a CNN's image classification.This study was registered at the German Clinical Trial Register (DRKS-Study-ID: DRKS00013570; https://www.drks.de/drks_web/).
|
|
3. |
- Tognetti, Linda, et al.
(author)
-
Development and Implementation of a Web-Based International Registry Dedicated to Atypical Pigmented Skin Lesions of the Face: Teledermatologic Investigation on Epidemiology and Risk Factors.
- 2023
-
In: Telemedicine journal and e-health : the official journal of the American Telemedicine Association. - : Mary Ann Liebert Inc. - 1556-3669. ; 29:9, s. 1356-1365
-
Journal article (peer-reviewed)abstract
- Background: Atypical pigmented facial lesions (aPFLs) often display clinical and dermoscopic equivocal and/or overlapping features, thus causing a challenging and delayed diagnosis and/or inappropriate excisions. No specific registry dedicated to aPFL paired with clinical data is available to date. Methods: The dataset is hosted on a specifically designed web platform. Each complete case was composed of the following data: (1) one dermoscopic picture; (2) one clinical picture; (3) two lesion data, that is, maximum diameter and facial location (e.g., orbital area/forehead/nose/cheek/chin/mouth); (4) patient's demographics: family history of melanoma, history of sunburns in childhood, phototype, pheomelanine, eyes/hair color, multiple nevi/dysplastic nevi on the body; and (5) acquisition device (videodermatoscope/camera-based/smartphone-based system). Results: A total of 11 dermatologic centers contributed to a final teledermoscopy database of 1,197 aPFL with a distribution of 353 lentigo maligna (LM), 146 lentigo maligna melanoma (LMM), 231 pigmented actinic keratoses, 266 solar lentigo, 125 atypical nevi, 48 seborrheic keratosis, and 28 seborrheic-lichenoid keratoses. The cheek site was involved in half of aPFL cases (50%). Compared with those with the other aPFL cases, patients with LM/LMM were predominantly men, older (69.32±12.9 years on average vs. 62.69±14.51), exhibited larger lesions (11.88±7.74mm average maximum diameter vs. 9.33±6.46mm), and reported a positive history of sunburn in childhood. Conclusions: The iDScore facial dataset currently represents a precious source of data suitable for the design of diagnostic support tools based on risk scoring classifiers to help dermatologists in recognizing LM/LMM among challenging aPFL in clinical practice.
|
|