| 2. |
- Tschandl, P., et al.
(författare)
-
Comparison of the accuracy of human readers versus machine-learning algorithms for pigmented skin lesion classification: an open, web-based, international, diagnostic study
- 2019
-
Ingår i: The Lancet Oncology. - 1470-2045 .- 1474-5488. ; 20:7, s. 938-947
-
Tidskriftsartikel (refereegranskat)abstract
- Background: Whether machine-learning algorithms can diagnose all pigmented skin lesions as accurately as human experts is unclear. The aim of this study was to compare the diagnostic accuracy of state-of-the-art machine-learning algorithms with human readers for all clinically relevant types of benign and malignant pigmented skin lesions. Methods: For this open, web-based, international, diagnostic study, human readers were asked to diagnose dermatoscopic images selected randomly in 30-image batches from a test set of 1511 images. The diagnoses from human readers were compared with those of 139 algorithms created by 77 machine-learning labs, who participated in the International Skin Imaging Collaboration 2018 challenge and received a training set of 10 015 images in advance. The ground truth of each lesion fell into one of seven predefined disease categories: intraepithelial carcinoma including actinic keratoses and Bowen's disease; basal cell carcinoma; benign keratinocytic lesions including solar lentigo, seborrheic keratosis and lichen planus-like keratosis; dermatofibroma; melanoma; melanocytic nevus; and vascular lesions. The two main outcomes were the differences in the number of correct specific diagnoses per batch between all human readers and the top three algorithms, and between human experts and the top three algorithms. Findings: Between Aug 4, 2018, and Sept 30, 2018, 511 human readers from 63 countries had at least one attempt in the reader study. 283 (55·4%) of 511 human readers were board-certified dermatologists, 118 (23·1%) were dermatology residents, and 83 (16·2%) were general practitioners. When comparing all human readers with all machine-learning algorithms, the algorithms achieved a mean of 2·01 (95% CI 1·97 to 2·04; p<0·0001) more correct diagnoses (17·91 [SD 3·42] vs 19·92 [4·27]). 27 human experts with more than 10 years of experience achieved a mean of 18·78 (SD 3·15) correct answers, compared with 25·43 (1·95) correct answers for the top three machine algorithms (mean difference 6·65, 95% CI 6·06–7·25; p<0·0001). The difference between human experts and the top three algorithms was significantly lower for images in the test set that were collected from sources not included in the training set (human underperformance of 11·4%, 95% CI 9·9–12·9 vs 3·6%, 0·8–6·3; p<0·0001). Interpretation: State-of-the-art machine-learning classifiers outperformed human experts in the diagnosis of pigmented skin lesions and should have a more important role in clinical practice. However, a possible limitation of these algorithms is their decreased performance for out-of-distribution images, which should be addressed in future research. Funding: None. © 2019 Elsevier Ltd
|
|
| 3. |
- Zaar, Oscar, et al.
(författare)
-
Dermoscopy of porokeratosis: results from a multicentre study of the International Dermoscopy Society
- 2021
-
Ingår i: Journal of the European Academy of Dermatology and Venereology. - : Wiley. - 0926-9959 .- 1468-3083. ; 35:10, s. 2091-2096
-
Tidskriftsartikel (refereegranskat)abstract
- Background The diagnosis of porokeratosis can be challenging, and knowledge about its dermoscopic features is limited. Objectives To describe the dermoscopic features of porokeratosis of Mibelli and disseminated superficial actinic porokeratosis (DSAP) and the frequency of these features in a larger case series. The interobserver concordance was also assessed. Methods In this retrospective cohort study, members of the International Dermoscopy Society contributed macroscopic and dermoscopic images of histopathologically verified cases of porokeratosis of Mibelli or DSAP. Three observers independently reviewed the collected images to identify the presence of predefined dermoscopic features. Following this, a consensus meeting was held to agree upon which dermoscopic features were present in each lesion. Results In total, 78 clinical and dermoscopic images of porokeratoses were collected. The most common dermoscopic feature was keratin rim, which was present in 74 lesions (92.3%). The most common vascular structures were dotted or glomerular vessels which were present in almost half of the cases (48.7%). Other relatively frequent dermoscopic findings were as follows: non-peripheral scales (44.9%), grey-brown dots or pigmentation along the keratin rim (38.5%), and light-brown pigmentation within the keratin rim (33.3%). Shiny white structures and blood spots or erosions along the keratin rim were findings never before described in porokeratosis and were detected in 16.7% and 17.9% of the lesions, respectively. Dermoscopic findings in porokeratosis of Mibelli and DSAP were similar except for fewer blood spots or erosions along the keratin rim and more light-brown pigmentation within the keratin rim in DSAP. The interobserver concordance ranged from 0.44 (moderate) to 0.84 (almost perfect). Conclusions The dermoscopic hallmark of porokeratosis is the keratin rim, a finding also allowing for almost perfect interobserver agreement. Pigmentation or erosions along the keratin rim, vascular structures, as well as scales, pigmentation or shiny white structures within the keratin rim are additional dermoscopic clues.
|
|
