Classification of Oral Cancer into Precancerous Stages from White images Using Light GBM Algorithm

Oral cancer is a significant global health concern, requiring precise and early diagnosis to improve treatment outcomes. This project enhances oral cancer classification accuracy using CNNs and K-means clustering-based image segmentation. The methodology involves preprocessing images, segmenting using K-means, and classifying into leukoplakia, erythroleukoplakia, and erythroplakias. Traditional approaches rely on visual examination and biopsy, which may lead to delays and misdiagnosis. The proposed system uses image processing and deep learning techniques to classify oral cancer into precancerous stages. The system achieves reliable and accurate classification, assisting medical professionals in early diagnosis and improving patient care.