Bone Tumour detection Using Feature Extraction with Classification by Deep Learning Techniques

Abstract

The majority of early deaths worldwide are attributed to a group of diseases known as bone cancers, which are characterised by unchecked cell development. This research propose novel technique in detection of bone cancer based on feature extraction and classification using deep learning architectures. Here the input has been processed and segmented for noise removal, image resize and smoothening. The features has been extracted using Convolutional histogram of oriented gradients (CHOG) and ROI extraction is used to improve the accuracy identification of abnormal part around the affected region. Then to classify the exact spotting and to grade the bone tissue as normal and abnormal using extreme Convolutional Deep learning machine (ECDLM). The results of the classification performance show that the neural network has a 92.50% success rate in classifying bone tumours.