Notably, the AUROCs were between 85% and 91% for artificial intelligence assessments of images from all 3 cameras with respect to biopsied lesions and between 91% and 96% for all lesions. The corresponding AUROC for biopsied lesions made by clinician assessment was 77.8%.
Furthermore, at a sensitivity of 100%, the specificity of the algorithm was 64.8% compared with 69.8% for clinician assessment.
In their concluding remarks, the study authors noted that “the findings of this diagnostic trial demonstrated that an artificial intelligence algorithm, using different camera types, can detect melanoma with a similar level of accuracy as specialists. The development of low-cost screening methods, such as artificial intelligence-based services, could transform patient diagnosis pathways, enabling greater efficiencies throughout the health care service.”
1. Phillips M, Marsden H, Jaffe W, et al. Assessment of accuracy of an artificial intelligence algorithm to detect melanoma in images of skin lesions. JAMA Netw Open. doi: 10.1001/jamanetworkopen.2019.13436
2. US Preventive Services Task Force, Bibbins-Domingo K, Grossman DC, Curry SJ, et al. Screening for skin cancer: US Preventive Services Task Force recommendation statement. JAMA. 2016;316(4):429-435.