• Animita Das
• Dr.Tabitha Janumala

Scoliosis is defined as a sideways curvature of the vertebral column. The current clinical approach uses radiographic images to determine the type of spine deformity and the severity of scoliosis is assessed by Cobb’s Angle. There are different types of scoliosis based on the region of bend such as thoracic, thoracolumbar, lumbar, and double major. Classification of scoliosis is of utmost importance for guiding the treatment of scoliosis. This paper proposes an automated scoliosis classification system based on a multiclass SVM model that uses pretrained deep neural networks as a feature extractor and a multi svm model to classify the scoliotic images into three categories Thoracic, Thoracolumbar, and Lumbar. A comparison was done using three pre-trained DNN.Statistical measures such as accuracy, Precision, Recall, and F1score were calculated and tabulated for each of them for performance evaluation.

Cobb’s angle, Classification, Deep Neural Networks, Feature Extraction, Multiclass SVM model, Scoliosis

"Classification of Scoliosis based on Multiclass SVM Model", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.8, Issue 6, page no.a930-a935, June-2021, Available :http://www.jetir.org/papers/JETIR2106130.pdf

"Classification of Scoliosis based on Multiclass SVM Model", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.8, Issue 6, page no. ppa930-a935, June-2021, Available at : http://www.jetir.org/papers/JETIR2106130.pdf