Diagnosis of Scalp Disorders Using Machine Learning and Deep Learning Approach: A Review

Scalp disorders, though less severe than other diseases, significantly impact patients' lives, with around 70% of adults experiencing issues like dandruff, psoriasis, and alopecia. Early diagnosis and treatment can reverse hair quality impairments caused by these conditions. Recent advances in deep learning, particularly using CNN and FCN, have shown high effectiveness in diagnosing scalp disorders. A proposed system combining an imaging microscope and a trained model achieved classification precision between 97.41% and 99.09%. Another study classified psoriasis with 82.9% accuracy using CNN, while an ML-based algorithm distinguished healthy scalps from alopecia areata with 91.4% and 88.9% accuracy using SVM and KNN. Despite these advancements, there is still room for improvement in using deep learning for scalp disease diagnosis.