AI Boosts Skin Cancer Diagnoses, Study Says

WEDNESDAY, June 18, 2025 (HealthDay News) — An experimental AI tool can help speed detection of melanoma and other skin diseases, a new study says.

The tool, PanDerm, improved accuracy of skin cancer diagnoses by 11% when used by doctors, researchers reported recently in the journal Nature Medicine.

It also helped doctors improve by nearly 17% accurate diagnosis of other skin conditions, researchers said.

“PanDerm is a tool designed to work alongside clinicians, helping them interpret complex imaging data and make informed decisions with more confidence,” senior researcher Zongyuan Ge said in a news release. He’s an associate professor of data science and AI at Monash University in Australia.

An estimated 70% of people have some sort of skin condition, researchers said in background notes. This makes early detection crucial, as it can lead to better treatment.

PanDerm was developed by an international team of researchers and trained on more than 2 million skin images, across four types of medical imaging, researchers said.

The AI was specifically designed to look for many different types of skin conditions, using images ranging from microscopic slides to wide-field images showing lesions and surrounding skin.

“Previous AI models have struggled to integrate and process various data types and imaging methods, reducing their usefulness to doctors in different real-world settings,” Ge said.

For this study, researchers tested the AI on a range of clinical tasks involving skin – cancer screening, mole counting, lesion changes and diagnosing a wide-ranging skin conditions.

PanDerm consistently delivered solid results, often with just 5% to 10% of the data normally required for diagnoses, results show.

"By training PanDerm on diverse data from different imaging techniques, we've created a system that can understand skin conditions the way dermatologists do — by synthesizing information from various visual sources," lead researcher Siyuan Yan, a doctoral student at the Monash University Faculty of Engineering, said in a news release.

This AI could be particularly important in places where access to dermatologists is limited, researcher Peter Soyer, director of the University of Queensland Dermatology Research Center in Australia, said in a news release.

However, PanDerm needs additional evaluation before it can be approved and put into use, researchers said.

The team plans to further study its performance in real-world settings, particularly across different health care environments treating different types of patients.

More information

The Cleveland Clinic has more on skin diseases.

SOURCE: Monash University, news release, June 6, 2025