Classification of Thyroid Carcinoma using Sobel Edge Detection and Adaptive Neuro Fuzzy Inference System Approach

Bioinformatics applications in the medical world become human needs. Pathologists as analysts required computer-assisted to conduct a diagnosis. This research aims to create an application that could help pathologic to diagnose thyroid cancer. The diagnosis could recognize the types of papillary, follicular, and anaplastic thyroid cancers. The cancer image was extracted using Sobel edge detection to derive the numeric database value. The numeric value represented the number of cells, cell expenses, and features of malignant cells. Furthermore, the ANFIS classified hybrid was forward and backward. The backward stage would correct the forward value for processing image data. Train image and probe image would be validated by pathologist as an accuracy test. The results obtained had a quite high accuracy of about 80%. This research had still need to be continued through a combination of feature extraction, classification and optimization techniques for higher performance. This application required further development as the pathologist needs visually in enforcing the gold standard examination.