In this study, machine learning-based computer-aided diagnosis significantly improved accuracy in diagnosing alopecia areata compared to traditional visual methods, achieving up to 91.9% accuracy using different classifiers like CNN, SVM, and random forest models.
1 citations
,
August 2023 in “arXiv (Cornell University)” This study reports that deep learning models, particularly CNN and FCN, achieved high accuracy in diagnosing scalp and skin disorders, suggesting potential for improved diagnostic systems with further advancements.
79 citations
,
July 2022 in “Sensors” In this study, researchers evaluated various machine learning models for predicting type 2 diabetes risk, finding that Random Forest and K-NN models performed best in terms of precision, recall, accuracy, and other metrics using common symptoms as features.
April 2019 in “Journal of Investigative Dermatology” This study reported that mSKPs and DMSCs share similarities in biological characteristics but exhibit distinct transcriptome profiles, with mSKPs being more immune-related and DMSCs more associated with differentiation and disease pathways.
May 2026 in “International Journal of Drug Delivery Technology” This study reports that using machine learning models, particularly XGBoost and Random Forest, can accurately predict PCOS phenotypes based on non-invasive data, with cycle length as the most significant predictor.
