Classification of Visually Similar Scalp Diseases Using Deep Learning: A Hybrid CNN-VIT Approach With Cross-Attention Fusion
June 2026
in “
Zenodo (CERN European Organization for Nuclear Research)
”
New to Alopecia Areata? There is a guide in the encyclopedia. Read the guide → 
Studysummary In this study, researchers developed a hybrid deep learning model called ScalpViT that accurately diagnosed scalp diseases with 94.3% accuracy, surpassing existing methods like ResNet-50 and EfficientNet-B3, and providing visual explainability for clinicians using GradCAM and Attention Rollout techniques.
Our plain-language summary. Not medical advice or a treatment recommendation. Consult a qualified healthcare professional before changing treatment. Full disclaimer
The study introduces ScalpViT, a novel hybrid deep learning model designed to accurately diagnose visually similar scalp diseases, such as Psoriasis, Seborrheic Dermatitis, Tinea Capitis, Alopecia Areata, Folliculitis, and Eczema. These conditions often share overlapping visual features, complicating diagnosis. ScalpViT combines a Vision Transformer (ViT) and a Convolutional Neural Network (CNN) with a cross-attention fusion module to enhance diagnostic accuracy. Trained on a dataset of approximately 7,000 images, ScalpViT achieved a 94.3% accuracy rate, outperforming existing models like ResNet-50 and EfficientNet-B3. The model also provides visual explainability through GradCAM and Attention Rollout, aiding clinical interpretation and deployment, especially in resource-limited settings.
