IMPORTANCE A high proportion of suspicious pigmented skin lesions referred for investigation are benign. Techniques to improve the accuracy of melanoma diagnoses throughout the patient pathway are needed to reduce the pressure on secondary care and pathology services. OBJECTIVE To determine the accuracy of an artificial intelligence algorithm in identifying melanoma in dermoscopic images of lesions taken with smartphone and digital single-lens reflex (DSLR) cameras. DESIGN, SETTING, AND PARTICIPANTS This prospective, multicenter, single-arm, masked diagnostic trial took place in dermatology and plastic surgery clinics in 7 UK hospitals. Dermoscopic images of suspicious and control skin lesions from 514 patients with at least 1 suspicious pigmented skin lesion scheduled for biopsy were captured on 3 different cameras. Data were collected from January 2017 to July 2018. Clinicians and the Deep Ensemble for Recognition of Malignancy, a deterministic artificial intelligence algorithm trained to identify melanoma in dermoscopic images of pigmented skin lesions using deep learning techniques, assessed the likelihood of melanoma. Initial data analysis was conducted in September 2018; further analysis was conducted from February 2019 to August 2019. INTERVENTIONS Clinician and algorithmic assessment of melanoma.MAIN OUTCOMES AND MEASURES Area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity of the algorithmic and specialist assessment, determined using histopathology diagnosis as the criterion standard. RESULTSThe study population of 514 patients included 279 women (55.7%) and 484 white patients (96.8%), with a mean (SD) age of 52.1 (18.6) years. A total of 1550 images of skin lesions were included in the analysis (551 [35.6%] biopsied lesions; 999 [64.4%] control lesions); 286 images (18.6%) were used to train the algorithm, and a further 849 (54.8%) images were missing or unsuitable for analysis. Of the biopsied lesions that were assessed by the algorithm and specialists, 125 (22.7%) were diagnosed as melanoma. Of these, 77 (16.7%) were used for the primary analysis.The algorithm achieved an AUROC of 90.1% (95% CI, 86.3%-94.0%) for biopsied lesions and 95.8% (95% CI, 94.1%-97.6%) for all lesions using iPhone 6s images; an AUROC of 85.8% (95% CI, 81.0%-90.7%) for biopsied lesions and 93.8% (95% CI, 91.4%-96.2%) for all lesions using Galaxy S6 images; and an AUROC of 86.9% (95% CI, 80.8%-93.0%) for biopsied lesions and 91.8% (95% CI, 87.5%-96.1%) for all lesions using DSLR camera images. At 100% sensitivity, the algorithm achieved (continued) Key Points Question How accurate is an artificial intelligence-based melanoma detection algorithm, which analyzes dermoscopic images taken by smartphone and digital single-lens reflex cameras, compared with clinical assessment and histopathological diagnosis? Findings In this diagnostic study, 1550 images of suspicious and benign skin lesions were analyzed by an artificial intelligence algorithm. When compared with histopathological diagnosis, ...

Background: Malignant melanoma can most successfully be cured when diagnosed at an early stage in the natural history. However, there is controversy over screening programs and many advocate screening only for high-risk individuals. Objectives: This study aimed to evaluate the accuracy of an artificial intelligence neural network (Deep Ensemble for Recognition of Melanoma [DERM]) to identify malignant melanoma from dermoscopic images of pigmented skin lesions and to show how this compared to doctors’ performance assessed by meta-analysis. Methods: DERM was trained and tested using 7,102 dermoscopic images of both histologically confirmed melanoma (24%) and benign pigmented lesions (76%). A meta-analysis was conducted of studies examining the accuracy of naked-eye examination, with or without dermoscopy, by specialist and general physicians whose clinical diagnosis was compared to histopathology. The meta-analysis was based on evaluation of 32,226pigmented lesions including 3,277 histopathology-confirmed malignant melanoma cases. The receiver operating characteristic (ROC) curve was used to examine and compare the diagnostic accuracy. Results: DERM achieved a ROC area under the curve (AUC) of 0.93 (95% confidence interval: 0.92-0.94), and sensitivity and specificity of 85.0% and 85.3%, respectively. Avoidance of false-negative results is essential, so different decision thresholds were examined. At 95% sensitivity DERM achieved a specificity of 64.1% and at 95% specificity the sensitivity was 67%. The meta-analysis showed primary care physicians (10 studies) achieve an AUC of 0.83 (95% confidence interval: 0.79-0.86), with sensitivity and specificity of 79.9% and 70.9%; and dermatologists (92 studies) 0.91 (0.88-0.93), 87.5%, and 81.4%, respectively. Conclusions: DERM has the potential to be used as a decision support tool in primary care, by providing dermatologist-grade recommendation on the likelihood of malignant melanoma.

Squamous cell carcinoma (SCC) and basal cell carcinoma (BCC) are common types of nonmelanoma skin cancer (NMSC). The DERM-003 study was a prospective, multicentre, single-arm, masked study that aimed to demonstrate the effectiveness of an artificial intelligence-based digital health technology (AI-DHT) to identify SCC, BCC and premalignant conditions in dermoscopic images of suspicious skin lesions. Patients with at least one suspicious skin lesion that was suitable for photography were eligible. Each lesion was photographed with three smartphone cameras (iPhone 6S, iPhone 11 and Samsung 10) with a dermoscopic lens attached. Each image was assessed by the AI-DHT. A clinical diagnosis was made by the dermatologist, and histopathology results were obtained for biopsied lesions. The AI-DHT output was compared with the histopathology diagnosis, and the area under the receiver operating characteristic curve (AUROC) was calculated. Secondary endpoints included other diagnostic measures and assessment of premalignant lesions. Altogether, 572 patients (49.5% female, mean age 68.5 years, 96.9% Fitzpatrick skin types I–III) were recruited from four UK National Health Service trusts, providing images of 611 suspicious lesions. There were no exclusion criteria relating to skin type. In total, 592 lesions had images from all three cameras available; only one lesion had no images available. Altogether, 395 lesions had a histopathology result. Forty-seven biopsied lesions were diagnosed as SCC and 184 as BCC. The AUROCs for images taken by the iPhone 6S were 0.88 [95% confidence interval (CI) 0.83–0.93] for SCC and 0.87 (95% CI 0.84–0.91) for BCC. For Samsung 10, the AUROCs were 0.85 (95% CI 0.79–0.90) and 0.87 (95% CI 0.83–0.90), and for iPhone 11, they were 0.88 (95% CI 0.84–0.93) and 0.89 (95% CI 0.86–0.92) for SCC and BCC, respectively. Using images taken on iPhone 6S of biopsied-only lesions, the sensitivity rates in detecting SCC and BCC were 98% (95% CI 88–100) and 94% (95% CI 90–97), respectively; the specificity rates were 38% (95% CI 33–44) and 28% (95% CI 21–35), respectively; the positive predictive values were 17% (95% CI 16–19) and 59% (95% CI 56–61), respectively; and the NPVs were 99% (95% CI 95–100) and 81% (95% CI 70–89), respectively. All 12 lesions diagnosed as Bowen disease were classified correctly by the AI-DHT. Of the 61 lesions diagnosed as actinic keratosis, 52 were correctly classified of the 22 lesions diagnosed as dysplastic and 18 were correctly classified by the AI-DHT. All 16 lesions diagnosed as melanoma were classified as such by the AI-DHT. The AI-DHT has the potential to support the diagnosis of NMSC and premalignant lesions.

Squamous cell carcinoma (SCC) and basal cell carcinoma (BCC) are common types of nonmelanoma skin cancer (NMSC). The DERM-003 study was a prospective multicentre single-arm masked study that aimed to demonstrate the effectiveness of an artificial intelligence-based digital health technology (AI-DHT) to identify SCC, BCC and premalignant conditions in dermoscopic images of suspicious skin lesions. Patients with at least one suspicious skin lesion that was suitable for photography were eligible. Each lesion was photographed with three smartphone cameras (iPhone 6S, iPhone 11 and Samsung 10) with a dermoscopic lens attached. Each image was assessed by the AI-DHT. A clinical diagnosis was made by the dermatologist, and histopathology results were obtained for biopsied lesions. The AI-DHT output was compared with the histopathology diagnosis, and the area under the receiver operating characteristic curve (AUROC) was calculated. Secondary endpoints included other diagnostic measures and assessment of premalignant lesions. Altogether, 572 patients (49.5% women, mean age 68.5 years, 96.9% Fitzpatrick skin types I–III) were recruited from four UK National Health Service trusts, providing images of 611 suspicious lesions. There were no exclusion criteria relating to skin type. In total, 592 lesions had images from all three cameras available; only one lesion had no images available. Altogether, 395 lesions had a histopathology result. Forty-seven biopsied lesions were diagnosed as SCC and 184 as BCC. The AUROCs for images taken by the iPhone 6S was 0.88 [95% confidence interval (CI) 0.83–0.93] for SCC and 0.87 (95% CI 0.84–0.91) for BCC. For the Samsung 10, the AUROCs were 0.85 (95% CI 0.79–0.90) and 0.87 (95% CI 0.83–0.90), and for the iPhone 11, they were 0.88 (95% CI 0.84–0.93) and 0.89 (95% CI 0.86–0.92) for SCC and BCC, respectively. Using images taken on the iPhone 6S of biopsied only lesions, the sensitivity in detecting SCC and BCC was 98% (95% CI 88–100) and 94% (95% CI 90–97), respectively; the specificity was 38% (95% CI 33–44) and 28% (95% CI 21–35), respectively; the positive predictive value was 17% (95% CI 16–19) and 59% (95% CI 56–61), respectively; and the NPV was 99% (95% CI 95–100) and 81% (95% CI 70–89), respectively. All 12 lesions diagnosed as Bowen disease were classified correctly by the AI-DHT. Of the 61 lesions diagnosed as actinic keratosis, 52 were correctly classified of the 22 lesions diagnosed as dysplastic, and 18 were correctly classified by the AI-DHT. All 16 lesions diagnosed as melanoma were classified as such by the AI-DHT. The AI-DHT has the potential to support the diagnosis of NMSC and premalignant lesions.