Classification of Oral lichen Planus Using Random Forest Ensemble Technique.

Abstract

Oral Lichen Planus (OLP) is a long-lasting inflammatory disorder disturbing the mucous membranes, with potential to escalate the risk of oral cancer if undiagnosed or mismanaged. Accurate and timely diagnosis is essential for effective treatment and management. This study compares the performance of three machine learning classifiers—Decision Tree, k-Nearest Neighbors (k-NN), and Random Forest—in detecting OLP based on MRI image data. A dataset of 50 MRI images was analyzed, with each classifier calculated using metrics such as accuracy, precision, recall, and F1 score. Outcomes indicate that the Random Forest classifier achieved the highest accuracy (90%), precision (92%), and recall (88%), outperforming the Decision Tree and k-NN classifiers, which yielded accuracies of 84% and 78%, respectively. While the Decision Tree demonstrated reasonable balance between precision and recall, k-NN showed lower sensitivity in detecting true OLP cases. These findings suggest that ensemble methods like Random Forest may offer superior diagnostic accuracy for OLP detection, underscoring the potential for machine learning to enhance clinical decision-making in oral health.