Digital imaging: a diagnostic screening tool?

Abstract

Much interest has been shown in the possibility of using digital images to aid in the more rapid diagnosis of patients with dermatologic disease. A study was undertaken to test the efficacy of the "teledermatology" screening of referrals, and thereby to triage the patients to appropriate care. A retrospective comparison of digital and "face-to-face" diagnoses by two consultant dermatologists was made. Eighty-four images from 75 patients seen in general dermatology clinics were studied. The clinical diagnosis and treatment plan of the patients seen in the clinic by one dermatologist were then compared with those suggested on digital image only, as seen by the other dermatologist. The diagnostic correlation between the two observers was fair, with full agreement in 47 of 84 cases (56%) and partial agreement in a further 10 (12%). There was no agreement in 14 cases (16.7%); in a further 13 cases, the images and history were not sufficient to allow a diagnosis to be made. The image quality was considered to be poor in 18 cases, but in six of these a diagnosis was still attempted. It was recommended that 66 patients (88%) should be seen in the hospital setting. Thirty patients (40%) received treatment at their visit. Teledermatology is not likely to have a great impact on reducing waiting lists. It is possible that it may help to prioritize referrals from remote areas.