Bone osteosarcoma tumor classification

Abstract

Osteosarcoma is a malignant bone tumor that usually affects children and adolescents. Early detection of osteosarcoma tumors increases the likelihood of successful therapy. Manual identification of osteosarcoma requires highly skilled doctors. In this study, we attempt to create a model to automatically diagnose tumors into three categories; non-tumor, viable-tumor, and osteosarcoma tumor. The suggested methodology can help medical professionals identify tumors correctly and quickly. The proposed approach uses the gray level co-occurrence matrix (GLCM) to extract features for feature extraction and three different classifiers for tumor detection. The used classifier are XG-Boost, support vector machine (SVM), and K-nearest neighbors. Finally, ensemble voting is used by combining the predictions from these classifiers. The system achieves 91.8% accuracy.