IIUM Repository (IREP)

Segmentation of lumbar spine MRI images for stenosis detection using patch-based pixel classification neural network

Al Kafri, Ala S and Sudirman, Sud and Hussain, Abir Jaafar and Al-Jumeily, Dhiya A. and Fergus, Paul and Natalia, Friska and Meidia, Hira and Afriliana, Nunik and Sophian, Ali and Al-Jumaily, Mohammed and Al-Rashdan, Wasfi and Bashtawi, Mohammad (2018) Segmentation of lumbar spine MRI images for stenosis detection using patch-based pixel classification neural network. In: 2018 IEEE Congress on Evolutionary Computation, CEC 2018, 8 July - 13 July 2018, Rio de Janeiro; Brazil.

[img] PDF (cover page) - Published Version
Restricted to Repository staff only

Download (131kB) | Request a copy
[img]
Preview
PDF (schedule) - Published Version
Download (319kB) | Preview
[img]
Preview
PDF (conference article) - Published Version
Download (4MB) | Preview
[img] PDF (scopus) - Supplemental Material
Restricted to Repository staff only

Download (674kB) | Request a copy

Abstract

This paper addresses the central problem of automatic segmentation of lumbar spine Magnetic Resonance Imaging (MRI) images to delineate boundaries between the anterior arch and posterior arch of the lumbar spine. This is necessary to efficiently detect the occurrence of lumbar spinal stenosis as a leading cause of Chronic Lower Back Pain. A patchbased classification neural network consisting of convolutional and fully connected layers is used to classify and label pixels in MRI images. The classifier is trained using overlapping patches of size 25x25 pixels taken from a set of cropped axial-view T2- weighted MRI images of the bottom three intervertebral discs. A set of experiment is conducted to measure the performance of the classification network in segmenting the images when either all or each of the discs separately is used. Using pixel accuracy, mean accuracy, mean Intersection over Union (IoU), and frequency weighted IoU as the performance metrics we have shown that our approach produces better segmentation results than eleven other pixel classifiers. Furthermore, our experiment result also indicates that our approach produces more accurate delineation of all important boundaries and making it best suited for the subsequent stage of lumbar spinal stenosis detection.

Item Type: Conference or Workshop Item (Invited Papers)
Additional Information: 7258/69097
Uncontrolled Keywords: Patch-based classification neural network, lumbar spine MRI, lumbar spinal stenosis
Subjects: Q Science > QA Mathematics > QA76 Computer software
Q Science > QP Physiology
R Medicine > RD Surgery
T Technology > T Technology (General)
Kulliyyahs/Centres/Divisions/Institutes: Kulliyyah of Engineering
Kulliyyah of Engineering > Department of Mechatronics Engineering
Depositing User: Dr Ali Sophian
Date Deposited: 31 Dec 2018 16:00
Last Modified: 31 Dec 2018 16:00
URI: http://irep.iium.edu.my/id/eprint/69097

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year