Three dimensional level set based semiautomatic segmentation of atherosclerotic carotid artery wall volume using 3D ultrasound imaging

Md. Murad Hossain; Khalid AlMuhanna; Limin Zhao; Brajesh K. Lal; Siddhartha Sikdar

doi:10.1117/12.2043975

21 March 2014 Three dimensional level set based semiautomatic segmentation of atherosclerotic carotid artery wall volume using 3D ultrasound imaging

Md. Murad Hossain, Khalid AlMuhanna, Limin Zhao, Brajesh K. Lal, Siddhartha Sikdar

Author Affiliations +

Proceedings Volume 9034, Medical Imaging 2014: Image Processing; 90344B (2014) https://doi.org/10.1117/12.2043975
Event: SPIE Medical Imaging, 2014, San Diego, California, United States

Abstract

3D segmentation of carotid plaque from ultrasound (US) images is challenging due to image artifacts and poor boundary definition. Semiautomatic segmentation algorithms for calculating vessel wall volume (VWV) have been proposed for the common carotid artery (CCA) but they have not been applied on plaques in the internal carotid artery (ICA). In this work, we describe a 3D segmentation algorithm that is robust to shadowing and missing boundaries. Our algorithm uses distance regularized level set method with edge and region based energy to segment the adventitial wall boundary (AWB) and lumen-intima boundary (LIB) of plaques in the CCA, ICA and external carotid artery (ECA). The algorithm is initialized by manually placing points on the boundary of a subset of transverse slices with an interslice distance of 4mm. We propose a novel user defined stopping surface based energy to prevent leaking of evolving surface across poorly defined boundaries. Validation was performed against manual segmentation using 3D US volumes acquired from five asymptomatic patients with carotid stenosis using a linear 4D probe. A pseudo gold-standard boundary was formed from manual segmentation by three observers. The Dice similarity coefficient (DSC), Hausdor distance (HD) and modified HD (MHD) were used to compare the algorithm results against the pseudo gold-standard on 1205 cross sectional slices of 5 3D US image sets. The algorithm showed good agreement with the pseudo gold standard boundary with mean DSC of 93.3% (AWB) and 89.82% (LIB); mean MHD of 0.34 mm (AWB) and 0.24 mm (LIB); mean HD of 1.27 mm (AWB) and 0.72 mm (LIB). The proposed 3D semiautomatic segmentation is the first step towards full characterization of 3D plaque progression and longitudinal monitoring.

Citation Download Citation

Md. Murad Hossain, Khalid AlMuhanna, Limin Zhao, Brajesh K. Lal, and Siddhartha Sikdar "Three dimensional level set based semiautomatic segmentation of atherosclerotic carotid artery wall volume using 3D ultrasound imaging", Proc. SPIE 9034, Medical Imaging 2014: Image Processing, 90344B (21 March 2014); https://doi.org/10.1117/12.2043975

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available