Enhanced Hair Disease Classification Using Deep Learning

This study presents a machine-learning model for classifying various hair conditions, achieving high performance with an overall accuracy of 97%. Using a dataset of 8,550 images, the model demonstrated precision, recall, and F1-Score metrics between 87.80% and 89.19%, with a total accuracy of 88.53%. The balanced F1 scores ranged from 88.02% to 88.99%, indicating a good trade-off between precision and recall. The macro and weighted average F1-Scores were 88.53% and 88.54%, respectively, highlighting the model's consistent and reliable performance across different hair disease categories.