Artificial intelligence is getting better at spotting trouble before it starts—especially when it comes to your skin. A new tool called PanDerm, developed by researchers led by Monash University, could give doctors an edge in detecting skin cancer early and accurately. Published in Nature Medicine, the study introduces one of the first AI systems designed to mirror how dermatologists actually work—by looking at different types of images together, not in isolation.
PanDerm is trained on more than two million skin images from 11 institutions across several countries. It’s not just looking at one kind of image. It analyses close-up photos, dermoscopic magnifications, full-body photographs, and even pathology slides. This approach makes it one of the few AI models built specifically to support real-world medical practice in dermatology.
Associate Professor Zongyuan Ge from Monash University’s Faculty of IT, one of the lead authors, explained that most existing AI models have been narrow in focus—useful for detecting specific cancers in specific kinds of images, but not much more. PanDerm, by contrast, works across formats. “Previous models have struggled to integrate and process various data types, reducing their usefulness to doctors in real-world settings,” he said. “PanDerm is a tool designed to work alongside clinicians, helping them interpret complex imaging data and make informed decisions with more confidence.”
The difference shows in the numbers. Doctors using PanDerm improved their skin cancer diagnosis accuracy by 11 per cent. The improvement was even greater for general practitioners and other non-specialists, who saw a 16.5 per cent boost across a range of skin conditions.
But it’s not just about percentages. PanDerm could help detect problems before they become visible to the human eye. The AI showed signs of picking up early changes in lesions—those subtle differences that can indicate something more serious may be developing.
Siyuan Yan, a PhD student from Monash Engineering and first author on the paper, said the key was training the AI to look at skin the way a doctor does: as a puzzle made up of many kinds of information. “By training PanDerm on diverse data from different imaging techniques, we’ve created a system that can understand skin conditions the way dermatologists do,” he said.
That matters, because access to specialists isn’t equal across the board. Whether you live in inner Melbourne or a rural Queensland town, the challenge is the same: spotting cancer early. Professor Victoria Mar, Director of the Alfred Health Victorian Melanoma Service, said PanDerm could assist doctors in keeping an eye on lesion changes over time and help predict which might pose a future risk. “This kind of assistance could support earlier diagnosis and more consistent monitoring for patients at risk of melanoma,” she said.
The tool has performed strongly in testing even with minimal labelled data—something researchers say is crucial for under-resourced settings. That includes parts of Australia where access to dermatologists is limited or inconsistent. Professor H. Peter Soyer from the University of Queensland said PanDerm’s strength lies in its ability to support current clinical routines rather than replace them. “It could be particularly valuable in busy or resource-limited settings, or in primary care where access to dermatologists may be limited,” he said.
Importantly, PanDerm is still in its evaluation phase. It’s not yet rolled out across clinics and hospitals, but the early signals are promising. The developers aim to refine how the system works across patient groups, settings, and conditions. That includes more thorough testing across different demographics to ensure fair and accurate performance regardless of age, skin colour, or geography.
Professor Harald Kittler, from the Medical University of Vienna, noted the power of international cooperation in developing PanDerm. “Its ability to support diagnosis in varied real-world settings, including in Europe, is a step forward in making dermatological expertise more accessible and consistent worldwide,” he said.
As AI tools begin to be taken seriously in frontline care, the key lies in trust, reliability, and integration. PanDerm’s approach—combining multiple image types, working with limited labelled data, and aligning with how doctors already operate—might just offer a credible blueprint for what future clinical AI should look like.
For now, the researchers say they are focusing on stress-testing the model further before it’s ready for broader deployment. But one thing is clear: your skin may soon have another pair of eyes watching out for it—and they won’t blink.
Support independent community journalism. Support The Indian Sun.
Follow The Indian Sun on X | Instagram | Facebook
🔍@MonashUni's #PanDermAI boosts #skincancer detection by 11-16.5% using multi-image analysis. 🏥 Trained on 2M+ global #skin images, it mimics #dermatologists' workflow. Aims to improve early diagnosis, especially in remote areas. #TheIndianSunhttps://t.co/4AwWyrSq1f
— The Indian Sun (@The_Indian_Sun) June 8, 2025
Donate To The Indian Sun
Dear Reader,The Indian Sun is an independent organisation committed to community journalism. We have, through the years, been able to reach a wide audience especially with the growth of social media, where we also have a strong presence. With platforms such as YouTube videos, we have been able to engage in different forms of storytelling. However, the past few years, like many media organisations around the world, it has not been an easy path. We have a greater challenge. We believe community journalism is very important for a multicultural country like Australia. We’re not able to do everything, but we aim for some of the most interesting stories and journalism of quality. We call upon readers like you to support us and make any contribution. Do make a DONATION NOW so we can continue with the volume and quality journalism that we are able to practice.
Thank you for your support.
Best wishes,
Team The Indian Sun
