Computational Analysis of Vertebral Body for Compression Fracture Using Texture and Shape Features

Computational Analysis of Vertebral Body for Compression Fracture Using Texture and Shape Features

Adela Arpitha, Lalitha Rangarajan
DOI: 10.4018/IJCINI.20211001.oa21
Article PDF Download
Open access articles are freely available for download

Abstract

The primary goal in this paper is to automate radiological measurements of Vertebral Body (VB) in Magnetic Resonance Imaging (MRI) spinal scans. It starts by preprocessing the images, then detect and localize the VB regions, next segment and label VBs and finally classify each VB into three cases as being normal or fractured in case 1, benign or malignant in case 2 and normal, benign or malignant in case 3. The task is accomplished by extracting and combining distinct features of VB such as boundary, gray levels, shape and texture features using various Machine Learning techniques. The class balance deficit dataset towards normal and fractures is balanced by data augmentation which provides an enriched dataset for the learning system to perform precise differentiation between classes. On a clinical spine dataset, the method is tested and validated on 535 VBs for segmentation attaining an average accuracy 94.59% and on 315 VBs for classification with an average accuracy of 96.07% for case 1, 93.23% for case 2 and 92.3% for case 3.
Article Preview
Top

1. Introduction

Imaging presents a window to peep into the patient anatomy. It may be very well utilized clinically to diagnose and select treatment for any abnormalities, in research to understand and devise new medicines for any abnormalities, and for routing during surgery planning. In this modern age, medical images are analysed for various automated tasks such as annotation, localization, labelling, segmentation, registration and classification all of which have become popular research topics. Computer Aided Diagnosis (CAD) acts as a practical tool to analyze any particular anatomical structure of a medical image to make better decisions in clinical diagnosis of pathologies, fractures, tumors, abnormalities and also for further planning of treatment, surgeries, post-surgical assessment and so forth (Rajaei et al., 2012). Medical institutions and hospitals produce immense medical images on a daily basis. Processing these images manually firstly requires proficiency / prowess in the field and secondly is an expensive and time consuming task to the professionals. This calls for the need of computerized image processing in this field.

In layman terms, a fracture is a break or crack in the bone. The human spine is built up of 33 vertebrae, which begins at the base of the skull: 7 cervical (C1- C, C7), 12 thoracic (T1 - T12), 5 lumbar (L1 - L5), 5 sacral bones welded forming the sacrum and 4 coccygeal bones which merge into one as coccyx. Every adjacent vertebra from cervical to lumbar is cushioned with an intervertebral disc in-between aiding flexibility to the spine. A vertebra is an irregular bone with a complex structure, divided into anterior and posterior parts. The posterior part of the vertebra, called the vertebral arch, which faces the back of a person, is formed by pairs of laminae and pedicles. Arising from the vertebral arch are four articular processes, two transverse processes and one spinous process. The vertebral forearm providing a passage for the spinal canal encloses and protects it. Vertebral Body (VB), the anterior part of the vertebra, is a cylindrical shaped portion that lies in front and provides majority of the structural support and stability to the spine.

Most of the Vertebral Fractures (VF) occur mainly in the VB section caused due to osteoporosis, injuries, infections and tumors, causing it to compress and is termed as compression fracture. Loss of height is eventually the result of multiple compression fractures. Compression fractures usually occur in the thoracolumbar spine (middle-lower) spine. In adults, VFs are commonly considered to be an indicator for osteoporosis and in pediatric cases its usually trauma related. Spinal fractures not only increase the risk of having another spinal fracture, but also of having other low-impact fractures in other parts of the body. All fractures cause deformities but the converse is not true. In clinical terms, a vertebral compression fracture subordinate to osteoporosis might be identified as a benign fracture, while if it’s brought about by bone metastasis it is generally characterized as a malignant fracture. Both benign and malignant fractures may be identified with similar complaints but discriminating the etiology between them at a beginning stage is significant to decide the clinical course, treatment, and prognosis (Frighetto-Pereira et al., 2015a). Different instances of normal, benign and malignant VBs are shown in Figure 1.

Figure 1.

Lumbar VB instances of normal, benign and malignant VBs

IJCINI.20211001.oa21.f01

Complete Article List

Search this Journal:
Reset
Volume 18: 1 Issue (2024)
Volume 17: 1 Issue (2023)
Volume 16: 1 Issue (2022)
Volume 15: 4 Issues (2021)
Volume 14: 4 Issues (2020)
Volume 13: 4 Issues (2019)
Volume 12: 4 Issues (2018)
Volume 11: 4 Issues (2017)
Volume 10: 4 Issues (2016)
Volume 9: 4 Issues (2015)
Volume 8: 4 Issues (2014)
Volume 7: 4 Issues (2013)
Volume 6: 4 Issues (2012)
Volume 5: 4 Issues (2011)
Volume 4: 4 Issues (2010)
Volume 3: 4 Issues (2009)
Volume 2: 4 Issues (2008)
Volume 1: 4 Issues (2007)
View Complete Journal Contents Listing