A Hybrid Approach to Osteosarcoma Detection Using Densenet201-SVM Model

Osteosarcoma, also known as osteogenic sarcoma, is the most prevalent primary bone tumor in children and is one of the leading causes of cancer-related mortality. Incidence peaks in adolescence, particularly during a growth spurt. It can occur in any bone, but the majority of cases arise from the metaphyses of long bones, this study, focuses on the development of an automated model which helps in the detection of Osteosarcoma, The dataset used in this research was obtained by a team of clinical researchers from the University of Texas Southwestern Medical Center in Dallas, through the use of archived samples from 50 pediatric patients who received treatment at Children's Medical Center in Dallas between 1995 and 2015.the detection process is achieved by combining one of the pre-trained transfer learning models namely, Densenet201 for feature extraction, and classifying these features in order to detect Osteosarcoma tumor, the proposed model achieved a high performance and accuracy of 90.96%.