A fully automated spinal cord segmentation

Subramanya Jois, SP and Sridhar, H and Harish Kumar, JR (2019) A fully automated spinal cord segmentation. In: UNSPECIFIED, 26 November 2018, pp. 524-528.

PDF Glo_Con_Sig_Inf_Pro_524_2019.pdf - Published Version Restricted to Registered users only Download (2MB) | Request a copy

Abstract

Segmentation of the spinal cord region is an imperative step in the automated analysis of neurological ailments such as multiple sclerosis. Multiple studies demonstrated the connection between progression of neurological diseases and measurements identifying with spinal cord atrophy and changes to its structure. Segmentation of spinal cord region manually or semi-automatically, can be conflicting and tedious for large datasets. We present a novel automated method, that segments the spinal cord region, utilizing circular active discs and region growth algorithm. The proposed method is validated on the Visible Human Project dataset. The results with regards to sensitivity, specificity, accuracy, Jaccard index, and Dice coefficient were 97.23, 100, 99.76, 96.83, and 98.65, respectively. The results were observed to be highly precise in comparison to expert outlines. © 2018 IEEE.

Item Type:	Conference Paper
Publication:	2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings
Publisher:	Institute of Electrical and Electronics Engineers Inc.
Additional Information:	Copyright fpr this article belongs to Institute of Electrical and Electronics Engineers Inc.
Keywords:	Image segmentation; Large dataset; Neurology, Active discs; Automated analysis; Multiple sclerosis; Neurological disease; Neurological disorders; Region growth; Spinal cords; Visible human project, Automation
Department/Centre:	Division of Electrical Sciences > Electrical Engineering
Date Deposited:	06 May 2019 12:48
Last Modified:	06 May 2019 12:48
URI:	http://eprints.iisc.ac.in/id/eprint/62156

Actions (login required)

View Item