CNN-KNN Model for Assessing Hair Health in Telogen Effluvium

The abstract provides a concise overview of a study that focuses on predicting hair breakage levels using a CNN-KNN hybrid model. The study evaluates the model's performance across five Breakage Degrees (BDs) by utilizing precision, recall, and F1-score metrics. Among these degrees, BD2 stands out with precision, recall, and F1-score values of 94.20%, indicating the model's ability to accurately classify hair damage levels between 21% and 40%. The support values, which indicate the number and proportion of instances in each BD, demonstrate a well-balanced distribution across all categories, ensuring the model's robust training. Researchers used a dataset of 10,050 images to reach an impressive 94.02% accuracy in predicting hair breakage levels caused by Telogen Effluvium in the research. Moreover, the model achieves an impressive overall accuracy of 98%, showcasing its effectiveness in distinguishing between various degrees of hair breakage severity. This high accuracy level highlights the model's potential application in clinical settings for dermatological diagnostics and personalized treatment planning in hair care management. The abstract underscores the CNN-KNN hybrid approach as a promising methodology for automated hair health assessment, providing healthcare professionals with a dependable tool to improve diagnostic accuracy and treatment efficacy. By incorporating machine learning techniques, specifically convolutional and k-nearest neighbor algorithms, the study contributes to the advancement of computational methods in dermatology and cosmetology. The results suggest practical implications for enhancing patient care through early detection and intervention of hair-related conditions, ultimately supporting better health outcomes and quality of life for individuals affected by hair breakage concerns.