Deep Convolutional Extreme Learning Machine with AlexNet-Based Bone Cancer Classification Using Whole-Body Scan Images

Bone cancer is a disease class that is characterized by freed cell growth, and this is the main cause of earlier death across the globe. Thus, earlier bone cancer detection and classification are required for curing the patient. This disease is originated from bone and spreads throughout the body rapidly and thus affects the patients. Quick analysis and initial diagnosis of bone cancer for creating the possible chance of protecting patients from death. Depending on this, bone cancer detection survey utilizing several techniques in image processing and several issues have found some of the problem complexity; when there is no detection performed at the right time, problem complexity increases. Hearing is a significant human feeling. The environment surrounding us is perceived and any danger happened around us is warned. A novel approach to identify the normal and malignant tissues is presented in this paper with the specific features in a provided X-ray image. This method utilizes AlexNet for feature extraction of a single category, and Deep Convolutional Extreme Machine Learning is utilized to create a classifier called Deep Convolutional Extreme Learning Machine with AlexNet (DC-ELM + AlexNet). The proposed method is evaluated with three standard approaches namely Model-averaged Neural Network (avNNET), Inception V3, and MobileNets in terms of various parameters. As a result, it is found that the proposed DC-ELM + AlexNet realizes 97.04% of accuracy, 68.22% of sensitivity, 83.94% of specificity, 54.08% of F1-score and 68.16% of kappa score.