Neural Network based classification of Cervical Spine Images(MRI) using Texture Features

Abstract

Medical imaging is a process of creating visual pictures of the internal parts of the human body for diagnostic and treatment purposes. It is considered to designate the set of techniques that noninvasively produce images of the internal aspect of the body. It can replace the face of the health care industry and permitted the scientists and practitioners to learn more about the human body than ever before. Medical image modalities allow the physician to monitor the effectiveness of treatment and adjust protocols as necessary. In this paper, Magnetic Resonance Images (MRI) are taken for processing and is a type of noninvasive test that uses magnets and radio waves to create images of the  internal parts of  the  body.  It can find a variety of conditions of the  cervical spine as well as problems in the soft tissues within the  spinal column, such as the spinal cord, nerves, and disks. Taken four statistical feature of texture such as energy, entropy, contrast and homogeneity was calculated from Grey Level Co-occurrence  Matrix  (GLCM). Comparison of the texture features are done through the classification between normal and herniated cervical spine images.