Detection of Squamous Cell Carcinoma Through Smart Technologies

Abstract

Oral cancer is the sixth most cancer type that occurs common among population who have addiction habit of chewing gutka, overconsumption of alcohol, excessive smoking and improper care for health. Squamous cell carcinoma is the most common type of oral cancer that occurs in mucosa lining of oral cavity. The proposed work implements simple and smart techniques based on colour feature to detect and localise the squamous cell carcinoma of oral cavity, as malignant tumour occur by changing the colour texture of tissue. Diagnosing colour appearance of any organ in human body plays a major role to locate the difference between healthy and abnormal changes of skin due to some underlying cause. The experimental results of the work locate region of interest (ROI) by converting RGB (red, green and blue) colour space to HSV (hue, saturation and value) colour space. Since HSV are used more common for tracking objects using computer vision and helps human to interpret colour changes easily. Then localised ROI is segmented into binary mask, by varying pixel intensity through track bars manually. Separation of susceptible cancerous region becomes more robust by applying contours on ROI of an image. This work also focuses on noise removal, thus making prediction possible for a physician.