Karapinar Senturk, ZehraGuler, RecepOzcan, YunusGamsizkan, Mehmet2025-10-112025-10-1120251532-06261532-0634https://doi.org/10.1002/cpe.70220https://hdl.handle.net/20.500.12684/22131This study presents a machine learning-based approach for the automated classification of skin diseases, specifically targeting morphea and lichen sclerosus, using microscopic images. The proposed method involves a systematic workflow, including image preprocessing techniques such as resizing, Reinhard normalization, Gaussian filtering, and CLAHE histogram equalization to enhance image quality. Feature extraction was performed using Gray-Level Co-occurrence Matrix (GLCM) and histogram-based statistical methods, capturing texture and intensity characteristics of skin tissues. Several classification models, including Support Vector Machine (SVM), Artificial Neural Network (ANN), Decision Tree (DT), Random Forest (RF), K-Nearest Neighbors (K-NN), and Logistic Regression (LR), were evaluated using accuracy, precision, recall, and F1 score, with hyperparameter optimization via grid search. The experimental results revealed that the combined feature set (GLCM + Histogram) achieved the highest performance, with the RF and K-NN models yielding a 100% in all performance metrics, including accuracy, sensitivity, recall, and F1-score. The study introduces a novel approach by examining these two diseases simultaneously, offering a reliable tool to support dermatologists with accurate and quick diagnoses. Future work will focus on expanding the dataset, exploring advanced deep learning techniques, and integrating clinical metadata to enhance model generalizability.en10.1002/cpe.70220info:eu-repo/semantics/openAccessmachine learningskin diseasewhole slide imageArticle3721-222-s2.0-105012477232WOS:001561266200046Q1Q3