Automatic Interpretation of Carotid Intima-Media Thickness Videos Using Convolutional Neural Networks

Nima Tajbakhsh, Jae Y. Shin, R. Todd Hurst, Christopher B. Kendall, Jianming Liang

Research output: Chapter in Book/Report/Conference proceedingChapter

2 Citations (Scopus)

Abstract

Cardiovascular disease (CVD) is the leading cause of death in the United States, yet it is largely preventable. But a critical part of prevention is identification of at-risk persons before adverse events. For predicting individual CVD risk, carotid intima-media thickness (CIMT), a noninvasive ultrasonography method, has proven to be valuable. However, each CIMT examination includes several ultrasonography videos, and interpreting each CIMT video involves 3 operations: (i) selecting 3 end-diastolic ultrasonographic frames (EUFs) in the video, (ii) localizing a region of interest (ROI) in each selected frame, and (iii) tracing the lumen-intima interface and the media-adventitia interface in each ROI to measure CIMT. These operations are tedious, laborious, and time-consuming, hindering the widespread utilization of CIMT in clinical practice. We present a new system based on convolutional neural networks to automate the entire process of CIMT video interpretation. The suggested system achieves a mean absolute error of 23.4 μm with a standard deviation of 17.3 μm for intima-media thickness measurements. The ANOVA test also yields p-values around 0.50, suggesting that there is a lack of evidence to show a difference among the CIMT measurements made by our system and those made by 3 experts. Our results suggest that the proposed system is robust against variability in CIMT measurements made by different experts.

Original languageEnglish (US)
Title of host publicationDeep Learning for Medical Image Analysis
PublisherElsevier Inc.
Pages105-131
Number of pages27
ISBN (Electronic)9780128104095
ISBN (Print)9780128104088
DOIs
StatePublished - Jan 30 2017

Fingerprint

Thickness measurement
Ultrasonography
Neural networks
Analysis of variance (ANOVA)

Keywords

  • Cardiovascular disease risk stratification
  • Carotid intima-media thickness
  • Convolutional neural networks
  • Deep learning

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Tajbakhsh, N., Shin, J. Y., Hurst, R. T., Kendall, C. B., & Liang, J. (2017). Automatic Interpretation of Carotid Intima-Media Thickness Videos Using Convolutional Neural Networks. In Deep Learning for Medical Image Analysis (pp. 105-131). Elsevier Inc.. https://doi.org/10.1016/B978-0-12-810408-8.00007-9

Automatic Interpretation of Carotid Intima-Media Thickness Videos Using Convolutional Neural Networks. / Tajbakhsh, Nima; Shin, Jae Y.; Hurst, R. Todd; Kendall, Christopher B.; Liang, Jianming.

Deep Learning for Medical Image Analysis. Elsevier Inc., 2017. p. 105-131.

Research output: Chapter in Book/Report/Conference proceedingChapter

Tajbakhsh, N, Shin, JY, Hurst, RT, Kendall, CB & Liang, J 2017, Automatic Interpretation of Carotid Intima-Media Thickness Videos Using Convolutional Neural Networks. in Deep Learning for Medical Image Analysis. Elsevier Inc., pp. 105-131. https://doi.org/10.1016/B978-0-12-810408-8.00007-9
Tajbakhsh N, Shin JY, Hurst RT, Kendall CB, Liang J. Automatic Interpretation of Carotid Intima-Media Thickness Videos Using Convolutional Neural Networks. In Deep Learning for Medical Image Analysis. Elsevier Inc. 2017. p. 105-131 https://doi.org/10.1016/B978-0-12-810408-8.00007-9
Tajbakhsh, Nima ; Shin, Jae Y. ; Hurst, R. Todd ; Kendall, Christopher B. ; Liang, Jianming. / Automatic Interpretation of Carotid Intima-Media Thickness Videos Using Convolutional Neural Networks. Deep Learning for Medical Image Analysis. Elsevier Inc., 2017. pp. 105-131
@inbook{0c1f31a7716745f4a511c2aa39095f3d,
title = "Automatic Interpretation of Carotid Intima-Media Thickness Videos Using Convolutional Neural Networks",
abstract = "Cardiovascular disease (CVD) is the leading cause of death in the United States, yet it is largely preventable. But a critical part of prevention is identification of at-risk persons before adverse events. For predicting individual CVD risk, carotid intima-media thickness (CIMT), a noninvasive ultrasonography method, has proven to be valuable. However, each CIMT examination includes several ultrasonography videos, and interpreting each CIMT video involves 3 operations: (i) selecting 3 end-diastolic ultrasonographic frames (EUFs) in the video, (ii) localizing a region of interest (ROI) in each selected frame, and (iii) tracing the lumen-intima interface and the media-adventitia interface in each ROI to measure CIMT. These operations are tedious, laborious, and time-consuming, hindering the widespread utilization of CIMT in clinical practice. We present a new system based on convolutional neural networks to automate the entire process of CIMT video interpretation. The suggested system achieves a mean absolute error of 23.4 μm with a standard deviation of 17.3 μm for intima-media thickness measurements. The ANOVA test also yields p-values around 0.50, suggesting that there is a lack of evidence to show a difference among the CIMT measurements made by our system and those made by 3 experts. Our results suggest that the proposed system is robust against variability in CIMT measurements made by different experts.",
keywords = "Cardiovascular disease risk stratification, Carotid intima-media thickness, Convolutional neural networks, Deep learning",
author = "Nima Tajbakhsh and Shin, {Jae Y.} and Hurst, {R. Todd} and Kendall, {Christopher B.} and Jianming Liang",
year = "2017",
month = "1",
day = "30",
doi = "10.1016/B978-0-12-810408-8.00007-9",
language = "English (US)",
isbn = "9780128104088",
pages = "105--131",
booktitle = "Deep Learning for Medical Image Analysis",
publisher = "Elsevier Inc.",
address = "United States",

}

TY - CHAP

T1 - Automatic Interpretation of Carotid Intima-Media Thickness Videos Using Convolutional Neural Networks

AU - Tajbakhsh, Nima

AU - Shin, Jae Y.

AU - Hurst, R. Todd

AU - Kendall, Christopher B.

AU - Liang, Jianming

PY - 2017/1/30

Y1 - 2017/1/30

N2 - Cardiovascular disease (CVD) is the leading cause of death in the United States, yet it is largely preventable. But a critical part of prevention is identification of at-risk persons before adverse events. For predicting individual CVD risk, carotid intima-media thickness (CIMT), a noninvasive ultrasonography method, has proven to be valuable. However, each CIMT examination includes several ultrasonography videos, and interpreting each CIMT video involves 3 operations: (i) selecting 3 end-diastolic ultrasonographic frames (EUFs) in the video, (ii) localizing a region of interest (ROI) in each selected frame, and (iii) tracing the lumen-intima interface and the media-adventitia interface in each ROI to measure CIMT. These operations are tedious, laborious, and time-consuming, hindering the widespread utilization of CIMT in clinical practice. We present a new system based on convolutional neural networks to automate the entire process of CIMT video interpretation. The suggested system achieves a mean absolute error of 23.4 μm with a standard deviation of 17.3 μm for intima-media thickness measurements. The ANOVA test also yields p-values around 0.50, suggesting that there is a lack of evidence to show a difference among the CIMT measurements made by our system and those made by 3 experts. Our results suggest that the proposed system is robust against variability in CIMT measurements made by different experts.

AB - Cardiovascular disease (CVD) is the leading cause of death in the United States, yet it is largely preventable. But a critical part of prevention is identification of at-risk persons before adverse events. For predicting individual CVD risk, carotid intima-media thickness (CIMT), a noninvasive ultrasonography method, has proven to be valuable. However, each CIMT examination includes several ultrasonography videos, and interpreting each CIMT video involves 3 operations: (i) selecting 3 end-diastolic ultrasonographic frames (EUFs) in the video, (ii) localizing a region of interest (ROI) in each selected frame, and (iii) tracing the lumen-intima interface and the media-adventitia interface in each ROI to measure CIMT. These operations are tedious, laborious, and time-consuming, hindering the widespread utilization of CIMT in clinical practice. We present a new system based on convolutional neural networks to automate the entire process of CIMT video interpretation. The suggested system achieves a mean absolute error of 23.4 μm with a standard deviation of 17.3 μm for intima-media thickness measurements. The ANOVA test also yields p-values around 0.50, suggesting that there is a lack of evidence to show a difference among the CIMT measurements made by our system and those made by 3 experts. Our results suggest that the proposed system is robust against variability in CIMT measurements made by different experts.

KW - Cardiovascular disease risk stratification

KW - Carotid intima-media thickness

KW - Convolutional neural networks

KW - Deep learning

UR - http://www.scopus.com/inward/record.url?scp=85027170494&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027170494&partnerID=8YFLogxK

U2 - 10.1016/B978-0-12-810408-8.00007-9

DO - 10.1016/B978-0-12-810408-8.00007-9

M3 - Chapter

SN - 9780128104088

SP - 105

EP - 131

BT - Deep Learning for Medical Image Analysis

PB - Elsevier Inc.

ER -