Image Processing Approaches for Oral Cancer Detection in Color Images

Lips, two-thirds of the tongue and inner cheek lining are all common places for oral cancer to form. It can also arise in hard and soft palates, pharynx, and sinuses. Head and neck cancers are subtypes of this malignancy. Unless detected and treated early on, oral can be dangerous. Using microscopic biopsy images, the researchers were able to detect mouth cancer and non-cancerous lesions. Using cutting-edge techniques, it is possible to histologically diagnose oral lesions. For example, enhancing microscopic images involves transforming them from RGB to L*a*b color space, classifying the colors using k-means clustering, segmenting the nuclei, and obtaining and classifying their features.