A prospective cohort study found that an autonomous total body photography and dermoscopic imaging device produced dermoscopic image quality comparable to manual dermoscopy and reduced imaging time in patients with numerous skin lesions.
Published in JAMA Dermatology, the study enrolled 316 adults with atypical mole syndrome at two dermatology clinics in Spain. Investigators compared an autonomous imaging system (Deviskan) with handheld dermoscopy. Two independent dermatologists assessed image quality and diagnostic agreement. The autonomous system achieved a mean image quality score of 9.84 out of 10, compared with 9.44 for manual dermoscopy. General body images were in focus in 99% of cases.
Diagnostic classification agreement between the two methods was 91.6%. Most discrepancies involved small, benign pigmented lesions, particularly melanocytic nevi vs benign keratoses.
Mean imaging time using the autonomous system was 518.77 seconds, which included a 2-minute instructional video, compared with 541.82 seconds for the manual method. In a subgroup of patients with more than 40 lesions, the time difference was statistically significant.
Participants had at least 30 lesions greater than 3 mm in diameter. Most lesions were located on the trunk (74%), followed by the upper arms (15%) and lower limbs (10%).
In total, 19,323 total body photography images and 10,646 dermoscopic images were captured, covering 5,323 lesions. The most common diagnoses were melanocytic nevus (76.15%) and benign keratosis (20.36%). Malignant lesions were uncommon and included 9 melanomas (0.08%) and 7 basal cell carcinomas (0.07%).
The Deviskan system uses a robotic camera and software to autonomously capture clinical and dermoscopic images. It incorporates patient positioning guidance, lesion detection, and tracking.
Image quality did not vary significantly by body site, lesion type, or size. Acceptability exceeded 98% across locations.
Limitations included a low prevalence of malignant lesions (2%), lack of representation of Fitzpatrick skin phototypes V and VI, and the exclusion of patients with pacemakers or large tattoos. The authors concluded that the autonomous imaging system was not inferior to standard dermoscopy and may improve efficiency in patients with a high number of lesions.
Full author disclosures are available in the article.
