AI Tool DEXI Helps Dermatologists Assess Skin Cancer Risk in Lesions with High Accuracy, Research Finds

Mayo Clinic researchers have found that an artificial intelligence tool may provide dermatologists with additional information to help assess whether skin lesions are cancerous and inform their clinical decisions about whether to biopsy or monitor them. The findings were published in Mayo Clinic Proceedings.

The tool, called Dermoscopy Explainable Intelligence (DEXI), is part of an advanced imaging system that dermatologists use to capture and track skin lesions over time. DEXI is a deep learning tool trained on more than 66,000 annotated dermoscopy images. It assigns lesions a score from 0.0 to 10.0 based on image characteristics associated with cancer risk and uses explainable AI to identify features such as asymmetry, border and colour that contribute to the score.

Clinicians can cross-check this output against their own examination. The tool’s transparency is intended to support rather than replace clinical expertise. This may help build confidence in integrating AI insights into routine care without diminishing the role of human judgement.

“Prior to the study, I would review a patient’s DEXI score, but I wouldn’t know exactly what to do with the number,” says Jenny Link, MD, dermatologist at Mayo Clinic and the study’s lead author. “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.”

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 analysed 261 lesions with both a DEXI score and a diagnosis confirmed by tissue examination. They compared scores for non-cancerous and cancerous lesions, further dividing the cancerous lesions into melanoma and non-melanoma skin cancers.

The average DEXI score was 2.5 for non-cancerous lesions, compared with 7.8 for non-melanoma skin cancers and 6.0 for melanomas. Cancerous lesions had significantly higher DEXI scores than non-cancerous lesions. Using a threshold of 3.0, with scores of 3.0 or higher considered positive for potential skin cancer, DEXI correctly identified 93.4% of cancerous lesions and 75% of non-cancerous lesions.

“We identified 3.0 as a potential DEXI score threshold that could provide useful information when dermatologists are considering whether to monitor a lesion or perform a biopsy,” says Dr Link.

Skin cancer remains a significant global health concern. According to the International Agency for Research on Cancer, almost 1.5 million new cases of skin cancer were estimated worldwide in 2024, including nearly 340,000 new melanoma cases and almost 60,000 deaths from the disease. Early and accurate assessment of suspicious lesions is critical to improving outcomes, which is why tools that support dermatologists in decision-making continue to attract research interest.

The researchers also identified limitations. DEXI does not account for patient-specific factors, such as age, medications and a history of melanoma. The tool was also less effective at evaluating severely atypical nevi, or skin moles, and early evolving melanomas.

“Our study demonstrated that DEXI is a useful and valuable tool to support clinicians as they evaluate cancerous and non-cancerous 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,” says Dr Link.

The study forms part of ongoing efforts at Mayo Clinic to explore how advanced imaging and AI can assist in dermatology. Further independent validation in larger and more diverse patient groups will be needed before wider clinical adoption. The full paper, including a complete list of authors, disclosures and funding information, is available in Mayo Clinic Proceedings.