New Denoising Unsharp Masking Method for Improved Intima Media Thickness Measurements with Active Contour Segmentation

Asraf Mohamed Moubark, David M.J. Cowell, Sevan Harput, Steven Freear

Research output: Contribution to journalConference article

Abstract

The semi-automated balloon snake active contour (BSAC) based segmentations play a vital role in determining the intima-media thickness (IMT) for accessing the risk related to cardio vascular diseases (CVD). However, the speckle and clutter noise in the ultrasound B-mode images are known to interfere with the contour formation during segmentation. Both noise sources act as false external energy in BSAC and thus influence the resulting boundary definition. A large number of iterations are required for the BSAC to accurately detect the boundary and in the presence of high noise the segmentation algorithm can result in false detections. Thus in this work we have applied the new denoising unsharp masking (UM) method on human common carotid artery in order to reduce clutter noise in the B-mode image before the segmentation process takes place for faster and accurate IMT measurement. The resuts show the number of iterations needed for BSAC to settle on the final intima-media border is less with UM-DAS (100 iterations) compared to that without the denoising technique, DAS (200 iterations). Thus the proposed UM techniques is able to provide better results with less time in measuring the IMT compared to that using DAS.

Original languageEnglish
Article number8579733
JournalIEEE International Ultrasonics Symposium, IUS
Volume2018-January
DOIs
Publication statusPublished - 1 Jan 2018
Externally publishedYes
Event2018 IEEE International Ultrasonics Symposium, IUS 2018 - Kobe, Japan
Duration: 22 Oct 201825 Oct 2018

Fingerprint

masking
snakes
balloons
iteration
clutter
arteries
borders
energy

ASJC Scopus subject areas

  • Acoustics and Ultrasonics

Cite this

New Denoising Unsharp Masking Method for Improved Intima Media Thickness Measurements with Active Contour Segmentation. / Mohamed Moubark, Asraf; Cowell, David M.J.; Harput, Sevan; Freear, Steven.

In: IEEE International Ultrasonics Symposium, IUS, Vol. 2018-January, 8579733, 01.01.2018.

Research output: Contribution to journalConference article

@article{d655f43519ef412da4061fd978d147c5,
title = "New Denoising Unsharp Masking Method for Improved Intima Media Thickness Measurements with Active Contour Segmentation",
abstract = "The semi-automated balloon snake active contour (BSAC) based segmentations play a vital role in determining the intima-media thickness (IMT) for accessing the risk related to cardio vascular diseases (CVD). However, the speckle and clutter noise in the ultrasound B-mode images are known to interfere with the contour formation during segmentation. Both noise sources act as false external energy in BSAC and thus influence the resulting boundary definition. A large number of iterations are required for the BSAC to accurately detect the boundary and in the presence of high noise the segmentation algorithm can result in false detections. Thus in this work we have applied the new denoising unsharp masking (UM) method on human common carotid artery in order to reduce clutter noise in the B-mode image before the segmentation process takes place for faster and accurate IMT measurement. The resuts show the number of iterations needed for BSAC to settle on the final intima-media border is less with UM-DAS (100 iterations) compared to that without the denoising technique, DAS (200 iterations). Thus the proposed UM techniques is able to provide better results with less time in measuring the IMT compared to that using DAS.",
author = "{Mohamed Moubark}, Asraf and Cowell, {David M.J.} and Sevan Harput and Steven Freear",
year = "2018",
month = "1",
day = "1",
doi = "10.1109/ULTSYM.2018.8579733",
language = "English",
volume = "2018-January",
journal = "IEEE International Ultrasonics Symposium, IUS",
issn = "1948-5719",

}

TY - JOUR

T1 - New Denoising Unsharp Masking Method for Improved Intima Media Thickness Measurements with Active Contour Segmentation

AU - Mohamed Moubark, Asraf

AU - Cowell, David M.J.

AU - Harput, Sevan

AU - Freear, Steven

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The semi-automated balloon snake active contour (BSAC) based segmentations play a vital role in determining the intima-media thickness (IMT) for accessing the risk related to cardio vascular diseases (CVD). However, the speckle and clutter noise in the ultrasound B-mode images are known to interfere with the contour formation during segmentation. Both noise sources act as false external energy in BSAC and thus influence the resulting boundary definition. A large number of iterations are required for the BSAC to accurately detect the boundary and in the presence of high noise the segmentation algorithm can result in false detections. Thus in this work we have applied the new denoising unsharp masking (UM) method on human common carotid artery in order to reduce clutter noise in the B-mode image before the segmentation process takes place for faster and accurate IMT measurement. The resuts show the number of iterations needed for BSAC to settle on the final intima-media border is less with UM-DAS (100 iterations) compared to that without the denoising technique, DAS (200 iterations). Thus the proposed UM techniques is able to provide better results with less time in measuring the IMT compared to that using DAS.

AB - The semi-automated balloon snake active contour (BSAC) based segmentations play a vital role in determining the intima-media thickness (IMT) for accessing the risk related to cardio vascular diseases (CVD). However, the speckle and clutter noise in the ultrasound B-mode images are known to interfere with the contour formation during segmentation. Both noise sources act as false external energy in BSAC and thus influence the resulting boundary definition. A large number of iterations are required for the BSAC to accurately detect the boundary and in the presence of high noise the segmentation algorithm can result in false detections. Thus in this work we have applied the new denoising unsharp masking (UM) method on human common carotid artery in order to reduce clutter noise in the B-mode image before the segmentation process takes place for faster and accurate IMT measurement. The resuts show the number of iterations needed for BSAC to settle on the final intima-media border is less with UM-DAS (100 iterations) compared to that without the denoising technique, DAS (200 iterations). Thus the proposed UM techniques is able to provide better results with less time in measuring the IMT compared to that using DAS.

UR - http://www.scopus.com/inward/record.url?scp=85062505393&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062505393&partnerID=8YFLogxK

U2 - 10.1109/ULTSYM.2018.8579733

DO - 10.1109/ULTSYM.2018.8579733

M3 - Conference article

AN - SCOPUS:85062505393

VL - 2018-January

JO - IEEE International Ultrasonics Symposium, IUS

JF - IEEE International Ultrasonics Symposium, IUS

SN - 1948-5719

M1 - 8579733

ER -