Explainable AI Dermoscopy Tool Supports Risk Assessment for Skin Cancer

Skin lesions can be difficult to triage when benign and malignant findings share visual features, leaving clinicians uncertain about whether to recommend biopsy or continued monitoring. These decisions require balancing the procedural burden of biopsy with the potential consequences of delayed cancer diagnosis. To support this assessment, Mayo Clinic researchers have evaluated an AI tool designed to assist dermatologists in assessing lesion risk.

The technology, called Dermoscopy Explainable Intelligence (DEXI), was evaluated by Mayo Clinic researchers as part of an advanced imaging system used to capture and track skin lesions over time. DEXI assigns each lesion a cancer-risk score from 0.0 to 10.0. It also identifies image features such as asymmetry, border and color that contributed to the score.

DEXI is a deep-learning tool trained on more than 66,000 annotated dermoscopy images. Its explainable artificial intelligence approach shows clinicians which visual characteristics influenced the risk assessment, allowing dermatologists to consider both the score and associated features alongside their clinical evaluation.

The retrospective study included 178 patients who underwent total-body photography, sequential digital dermoscopy imaging, and subsequent biopsy of one or more skin lesions. Researchers analyzed 261 lesions with both a DEXI score and a tissue-confirmed diagnosis, comparing noncancerous lesions with melanoma and nonmelanoma skin cancers.

At a threshold of 3.0, with scores at or above this value considered positive for potential skin cancer, DEXI correctly identified 93.4% of cancerous lesions and 75% of noncancerous lesions. Scores also differed by diagnosis: the average was 2.5 for noncancerous lesions, compared with 7.8 for nonmelanoma skin cancers and 6.0 for melanomas.

The findings were published in Mayo Clinic Proceedings on October 21, 2026. The researchers identified 3.0 as a potential threshold to help inform decisions about monitoring or biopsy. They also noted that DEXI does not account for patient-specific factors such as age, medications and history of melanoma, and that it was less effective for severely atypical nevi, or skin moles, and early evolving melanomas.

“Prior to the study, I would review a patient's DEXI score, but I wouldn't know exactly what to do with the number. Now I can confidently use the score as another piece of data to inform my clinical judgment on whether I recommend a biopsy or monitoring,” said Jenny Link, M.D., a dermatologist at Mayo Clinic and the study's lead author.

“Our study demonstrated that DEXI is a useful and valuable tool to support clinicians as they evaluate cancerous and noncancerous skin lesions, but it should not be used in isolation. The score must be interpreted alongside clinical findings and patient-specific factors, and the tool requires further validation and refinement,” added Link.

Related Links
Mayo Clinic