Premium
Teledermoscopy images acquired in primary health care and hospital settings – a comparative study of image quality
Author(s) -
Dahlén Gyllencreutz J.,
Johansson Backman E.,
Terstappen K.,
Paoli J.
Publication year - 2018
Publication title -
journal of the european academy of dermatology and venereology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.655
H-Index - 107
eISSN - 1468-3083
pISSN - 0926-9959
DOI - 10.1111/jdv.14565
Subject(s) - medicine , primary care , quality (philosophy) , primary health care , health care , image quality , family medicine , medical emergency , medical physics , image (mathematics) , artificial intelligence , environmental health , philosophy , epistemology , computer science , economics , economic growth , population
Background The incidence of melanoma and non‐melanoma skin cancer is increasing, which has also lead to an increase in referrals between primary health care ( PHC ) and dermatology departments, putting a strain on healthcare services. Teledermoscopy ( TDS ) referrals from PHC can improve the triage process for patients with suspicious skin tumours, but the quality of the images included could potentially affect its usefulness. Objective To critically appraise the quality of the dermoscopic images of a smartphone TDS system, by comparing the TDS referral images with images of the same tumours acquired at the department of dermatology. Methods Two dermatologists rated the image quality of two image sets from 172 skin tumours separately. The dermatologists also decided on a main diagnosis, differential diagnoses and described the visible dermoscopic structures. Results The images acquired in PHC were rated as having slightly lower quality, but there was no significant difference. PHC images and dermatology images were of intermediate‐to‐high quality in 95.5%–97.7% and 96.5%–98.8%, respectively. There was no difference in agreement between the TDS diagnosis based on the two image sets with the final clinical or histopathological diagnosis. Most image pairs (81.4% and 83.7%) received the same main diagnosis by the two evaluators. When this was not the case, the most common reasons were poor focus, excessive pressure applied when acquiring the image or inadequate amount of zoom. Conclusion TDS performed in PHC with a smartphone‐based system does not seem to negatively affect the usefulness of TDS referrals. Thus, physicians at PHC do not necessarily need to be trained photographers to ensure adequate TDS image quality. Knowledge about technical difficulties could however be used when training PHC staff, to improve the image quality further.