Improved Identification of Potentially Dangerous Pigmented Skin Lesions by Computerized Image Analysis

Digital computer analysis of dermatoscopical images of 260 melanocytic skin lesions; perimeter/area ratio for the differentiation between malignant melanomas and melanocytic nevi

“…Geometric, structural and colour features of examined lesions can be objectively evaluated; some computers even determine the definite diagnosis. Several studies have already confirmed the efficacy of digital computer analysis of dermatoscopical images for the diagnosis of melanocytic skin lesions 6–10 . However, the clinical application of this method in daily practice has not been clarified.…”

Potential Impact of Advanced Clinical Information Technology on Cancer Care in 2015

“…In addition, the positive predictive value (Pap Classes III-V) of Papnet was slightly but not significantly better (55% vs. 51%) [31]. Over the course of the next 10 years, further development and ultimately routine usage of such computer-assisted diagnosis systems for a whole host of other cancer types such as in the field of dermatology [32] is sure to increase, with the main effect being an increase in availability of highly accurate diagnoses and a decrease in labor costs associated with such screening procedures.…”

Evaluation of a Program for the Automatic Dermoscopic Diagnosis of Melanoma in a General Dermatology Setting

“…Their potential benefits are incalculable, but there are considerable difficulties involved in their development and use in clinical practice 1 . It has been widely stated that their accuracy can achieve the same range as dermoscopic classification by experts or even that they can obtain better accuracy 1,2,6 . A few authors have recently found them to be far less accurate than initially thought, 7,8 suggesting that the promising initial results might have been a result of intentional or unintentional selection bias of tested images 8 .…”

“…The accuracy of automated programs for the computerized, dermoscopic diagnosis of melanoma has been found to be similar or superior to that of expert clinicians, 1,2 even in evidence‐based meta‐analysis 1 . Most of this research has been performed in pigmented skin lesion (PSL) units, where the number of diagnostically challenging PSLs is high.…”