Automatic segmentation of common carotid artery in transverse mode ultrasound images

J. R.Harish Kumar, Chandra Sekhar Seelamantula, Nikhil S. Narayan, Pina Marziliano

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

We consider the problem of carotid artery segmentation and develop an automated outlining technique based on the active disc formalism that we recently introduced. The outlining problem is posed as one of optimization of a locally defined contrast function with respect to the affine transformation parameters that characterize the active disc. It turns out that standard techniques based on gradient-descent minimization can be used to carry out the optimization, although more sophisticated optimizers could also be deployed. For the initialization, we use a matched filter with a template size chosen based on an estimate of the average size of the carotid artery. We report results of experimental validation on Brno university's signal processing (SP) lab database, which contains 971 transverse mode ultrasound images of the carotid artery. The images in the database are manually annotated using a circle with center and radius explicitly specified in pixels, which serves as the reference. The circular annotation is also a good match with the active disc template considered in this paper. The proposed method results in an average detection accuracy of 95.5% and an average Dice similarity measure of 87.36% and takes only a few seconds of processing time per image. Comparisons with other state-of-the-art techniques are also reported.

Original languageEnglish
Title of host publication2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings
PublisherIEEE Computer Society
Pages389-393
Number of pages5
Volume2016-August
ISBN (Electronic)9781467399616
DOIs
Publication statusPublished - 03-08-2016
Externally publishedYes
Event23rd IEEE International Conference on Image Processing, ICIP 2016 - Phoenix, United States
Duration: 25-09-201628-09-2016

Conference

Conference23rd IEEE International Conference on Image Processing, ICIP 2016
CountryUnited States
CityPhoenix
Period25-09-1628-09-16

Fingerprint

Ultrasonics
Matched filters
Signal processing
Pixels
Processing

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Kumar, J. R. H., Seelamantula, C. S., Narayan, N. S., & Marziliano, P. (2016). Automatic segmentation of common carotid artery in transverse mode ultrasound images. In 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings (Vol. 2016-August, pp. 389-393). [7532385] IEEE Computer Society. https://doi.org/10.1109/ICIP.2016.7532385
Kumar, J. R.Harish ; Seelamantula, Chandra Sekhar ; Narayan, Nikhil S. ; Marziliano, Pina. / Automatic segmentation of common carotid artery in transverse mode ultrasound images. 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. Vol. 2016-August IEEE Computer Society, 2016. pp. 389-393
@inproceedings{be684feb507746fbabc26965ec76cebd,
title = "Automatic segmentation of common carotid artery in transverse mode ultrasound images",
abstract = "We consider the problem of carotid artery segmentation and develop an automated outlining technique based on the active disc formalism that we recently introduced. The outlining problem is posed as one of optimization of a locally defined contrast function with respect to the affine transformation parameters that characterize the active disc. It turns out that standard techniques based on gradient-descent minimization can be used to carry out the optimization, although more sophisticated optimizers could also be deployed. For the initialization, we use a matched filter with a template size chosen based on an estimate of the average size of the carotid artery. We report results of experimental validation on Brno university's signal processing (SP) lab database, which contains 971 transverse mode ultrasound images of the carotid artery. The images in the database are manually annotated using a circle with center and radius explicitly specified in pixels, which serves as the reference. The circular annotation is also a good match with the active disc template considered in this paper. The proposed method results in an average detection accuracy of 95.5{\%} and an average Dice similarity measure of 87.36{\%} and takes only a few seconds of processing time per image. Comparisons with other state-of-the-art techniques are also reported.",
author = "Kumar, {J. R.Harish} and Seelamantula, {Chandra Sekhar} and Narayan, {Nikhil S.} and Pina Marziliano",
year = "2016",
month = "8",
day = "3",
doi = "10.1109/ICIP.2016.7532385",
language = "English",
volume = "2016-August",
pages = "389--393",
booktitle = "2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings",
publisher = "IEEE Computer Society",
address = "United States",

}

Kumar, JRH, Seelamantula, CS, Narayan, NS & Marziliano, P 2016, Automatic segmentation of common carotid artery in transverse mode ultrasound images. in 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. vol. 2016-August, 7532385, IEEE Computer Society, pp. 389-393, 23rd IEEE International Conference on Image Processing, ICIP 2016, Phoenix, United States, 25-09-16. https://doi.org/10.1109/ICIP.2016.7532385

Automatic segmentation of common carotid artery in transverse mode ultrasound images. / Kumar, J. R.Harish; Seelamantula, Chandra Sekhar; Narayan, Nikhil S.; Marziliano, Pina.

2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. Vol. 2016-August IEEE Computer Society, 2016. p. 389-393 7532385.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Automatic segmentation of common carotid artery in transverse mode ultrasound images

AU - Kumar, J. R.Harish

AU - Seelamantula, Chandra Sekhar

AU - Narayan, Nikhil S.

AU - Marziliano, Pina

PY - 2016/8/3

Y1 - 2016/8/3

N2 - We consider the problem of carotid artery segmentation and develop an automated outlining technique based on the active disc formalism that we recently introduced. The outlining problem is posed as one of optimization of a locally defined contrast function with respect to the affine transformation parameters that characterize the active disc. It turns out that standard techniques based on gradient-descent minimization can be used to carry out the optimization, although more sophisticated optimizers could also be deployed. For the initialization, we use a matched filter with a template size chosen based on an estimate of the average size of the carotid artery. We report results of experimental validation on Brno university's signal processing (SP) lab database, which contains 971 transverse mode ultrasound images of the carotid artery. The images in the database are manually annotated using a circle with center and radius explicitly specified in pixels, which serves as the reference. The circular annotation is also a good match with the active disc template considered in this paper. The proposed method results in an average detection accuracy of 95.5% and an average Dice similarity measure of 87.36% and takes only a few seconds of processing time per image. Comparisons with other state-of-the-art techniques are also reported.

AB - We consider the problem of carotid artery segmentation and develop an automated outlining technique based on the active disc formalism that we recently introduced. The outlining problem is posed as one of optimization of a locally defined contrast function with respect to the affine transformation parameters that characterize the active disc. It turns out that standard techniques based on gradient-descent minimization can be used to carry out the optimization, although more sophisticated optimizers could also be deployed. For the initialization, we use a matched filter with a template size chosen based on an estimate of the average size of the carotid artery. We report results of experimental validation on Brno university's signal processing (SP) lab database, which contains 971 transverse mode ultrasound images of the carotid artery. The images in the database are manually annotated using a circle with center and radius explicitly specified in pixels, which serves as the reference. The circular annotation is also a good match with the active disc template considered in this paper. The proposed method results in an average detection accuracy of 95.5% and an average Dice similarity measure of 87.36% and takes only a few seconds of processing time per image. Comparisons with other state-of-the-art techniques are also reported.

UR - http://www.scopus.com/inward/record.url?scp=85006826048&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006826048&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2016.7532385

DO - 10.1109/ICIP.2016.7532385

M3 - Conference contribution

VL - 2016-August

SP - 389

EP - 393

BT - 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings

PB - IEEE Computer Society

ER -

Kumar JRH, Seelamantula CS, Narayan NS, Marziliano P. Automatic segmentation of common carotid artery in transverse mode ultrasound images. In 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. Vol. 2016-August. IEEE Computer Society. 2016. p. 389-393. 7532385 https://doi.org/10.1109/ICIP.2016.7532385