research BaldGraphFormer: An Explainable Deep Learning Framework for Early Baldness Prediction by Integrating Swin Transformer, Graph Attention Networks and Clinical Features
In this study, the BaldGraphFormer framework, integrating visual and clinical data, outperformed unimodal baselines in early-stage androgenetic alopecia detection, achieving an F1-score of 97.62% and macro-average AUC of 0.992, suggesting its potential to support dermatological decision-making and early intervention.