An Approach for the Detection and Classification of Tumor Cells from Bone MRI Using Wavelet Transform and KNN Classifier
Eftekhar Hossain, Md. Farhad Hossain, Mohammad Anisur Rahaman · 2018
Bone cancer is one of the most dangerous and leading cause of early death around the globe. Therefore, early detection and classification of the bone tumor have become needed to cure the patient. This study uses a wavelet-based segmentation method for the detection of the bone tumor. The segmented bone tumor part is further processed for the classification purpose. In this study k-nearest neighbour (KNN) classifier is employed for the classification of bone tumor into benign and malignant class. A number of images are collected and gray level co-occurrence matrix (GLCM) features are extracted from these images for the creation of a learned classifier model. The obtained performance of the classification result exhibit that the KNN classifier provides 92.50% accuracy in bone tumor classification.