New AI can improve non-derma’s ability to diagnose skin conditions

##New AI Model PanDerm Improves Skin Cancer Diagnosis Accuracy

Highlights:
1. A new AI model called PanDerm can simultaneously analyze different types of skin images to enhance diagnostic accuracy for dermatology and non-dermatology professionals.

2. The model, developed by an international research team led by AI and machine learning experts from Monash University, was trained on over two million skin images from 11 institutions worldwide to perform various clinical tasks, including skin cancer detection, mole counting, and differential diagnosis of skin conditions.

3. Findings published in Nature Medicine show that PanDerm outperformed clinicians in detecting early-stage melanoma by 10% and raised the accuracy of dermatologists in diagnosing skin cancer from dermoscopic images by 11%. It also enhanced non-dermatologists’ ability to identify and differentiate skin conditions based on photos by 16.5%.

Summary:
The PanDerm AI model developed by researchers from Monash University demonstrates significant advancements in dermatology by improving diagnostic accuracy for various skin conditions, including early-stage melanoma detection. Trained on a large dataset of skin images, PanDerm outperformed clinicians in several clinical tasks and shows promise in addressing dermatological expertise gaps across healthcare settings.

Editorial content by NewsSIP AI