Computer vision techniques for Upper Aero-Digestive Tract tumor grading classification – Addressing pathological challenges

Abstract

5• A novel hybrid FR-PSO based SVM model is proposed for Upper Aero Digestive Tract Cancer grading classification . 5• Morphological analysis of benign and malignant tumours is carried out using image processing techniques . 5• Finding inter cellular Bridges to find the dysplasia type of tumours in UADT. 5• A Force Reconstructed PSO technique is used to fine-tune the SVM hyper plane for multiple class grading classification. Oral cancer is one of the common cancer types which scales higher in death rate every year. The connectivity of two different cavities like oral cavity and nasal cavity is known as Upper Aero-Digestive Tract (UADT). Both oral and nasal cavities consist of thirteen connecting sites from mouth to upper stomach. The traditional pathological analysis like manual microscopic review brings out major intra and interobserver variability problem. A new automated system is proposed using computer vision techniques to focus and analyse major pathological problems like intra and interobserver variability problem and mis-classification of dysplasia type of tumours. The morphological behaviour of biopsy tissue samples are analysed digitally with different sites of UADT and different cancerous and non-cancerous stages. The proposed technique will play a major role in assisting the manual pathology procedure for analysing the morphology of dysplasia type of tumours and classification of tumour gradings. A method is proposed which integrates an alternate process to find the morphology of dysplasia type tumours using different image processing techniques. A state-of-the-art Force Reconstructed Particle Swarm Optimization Based SVM is proposed for UADT oral cancer classification for ten different oral cavity sites. The proposed classification technique achieved 94 % accuracy.