A Machine Learning Algorithm Applied to Trichoscopy for Androgenic Alopecia Staging and Severity Assessment

The study used a machine learning model to analyze trichoscopic images for the staging and severity assessment of Androgenic Alopecia (AGA). The model was found to be effective in identifying patients with moderate-severe AGA who were initially diagnosed with mild AGA based on classical macroscopic scales. The model uses trichoscopic parameters such as the percentage of vellus hairs, single hair follicular units, and the number of empty follicles. The model's severity index could guide dermatologists in choosing the right treatment regimen and monitor response to therapy. The model performed well in classifying AGA in most cases, with high precision, accuracy, and recall. However, for patients with a probability estimate near 50%, the classification might be unreliable. The model is best applicable in mild and moderate stages of AGA. The study suggests that the model could be a valuable tool for dermatologists in managing AGA, especially in defining a more precise staging and prescribing a more appropriate therapy.