SE-ResNet based vulnerable plaque recognition in IVOCT images

Wenliu Qi; Sihui Du; Xiaoying Tang; Ancong Wang

doi:10.1145/3574198.3574209

SE-ResNet based vulnerable plaque recognition in IVOCT images

Wenliu Qi, Sihui Du, Xiaoying Tang, Ancong Wang^*

^*Corresponding author for this work

School of Medical and Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Acute coronary syndrome (ACS) caused by vulnerable plaques can lead to sudden death. The resolution of intravascular optical coherence tomography (IVOCT) is up to 10 μm, and it has become the first choice for vulnerable plaque recognition. However, it is time-consuming and burdensome for doctors to label vulnerable plaques manually. As a result, it is important to develop an automatic method for vulnerable plaque recognition in IVOCT images. This paper proposes a lightweight and real-time method to identify the main vulnerable plaque areas in IVOCT images. The accuracy rate, recall rate and overlap rate of this method on the test set are 84.8%, 90.1%, and 87.0% respectively, and the recognition quality is 87.2%. The results suggest that our method may assist doctors to recognize vulnerable plaque areas fast and accurately.

Original language	English
Title of host publication	ICBBE 2022 - Proceeding of 2022 9th International Conference on Biomedical and Bioinformatics Engineering
Publisher	Association for Computing Machinery
Pages	68-73
Number of pages	6
ISBN (Electronic)	9781450397223
DOIs	https://doi.org/10.1145/3574198.3574209
Publication status	Published - 10 Nov 2022
Event	9th International Conference on Biomedical and Bioinformatics Engineering, ICBBE 2022 - Virtual, Online, Japan Duration: 10 Nov 2022 → 13 Nov 2022

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	9th International Conference on Biomedical and Bioinformatics Engineering, ICBBE 2022
Country/Territory	Japan
City	Virtual, Online
Period	10/11/22 → 13/11/22

Keywords

Convolutional neural network
Intravascular optical coherence tomography
Vulnerable plaque
image recognition

Access to Document

10.1145/3574198.3574209

Cite this

@inproceedings{9c9f355bcde74b49a955af322892a07c,

title = "SE-ResNet based vulnerable plaque recognition in IVOCT images",

abstract = "Acute coronary syndrome (ACS) caused by vulnerable plaques can lead to sudden death. The resolution of intravascular optical coherence tomography (IVOCT) is up to 10 μm, and it has become the first choice for vulnerable plaque recognition. However, it is time-consuming and burdensome for doctors to label vulnerable plaques manually. As a result, it is important to develop an automatic method for vulnerable plaque recognition in IVOCT images. This paper proposes a lightweight and real-time method to identify the main vulnerable plaque areas in IVOCT images. The accuracy rate, recall rate and overlap rate of this method on the test set are 84.8%, 90.1%, and 87.0% respectively, and the recognition quality is 87.2%. The results suggest that our method may assist doctors to recognize vulnerable plaque areas fast and accurately.",

keywords = "Convolutional neural network, Intravascular optical coherence tomography, Vulnerable plaque, image recognition",

author = "Wenliu Qi and Sihui Du and Xiaoying Tang and Ancong Wang",

note = "Publisher Copyright: {\textcopyright} 2022 ACM.; 9th International Conference on Biomedical and Bioinformatics Engineering, ICBBE 2022 ; Conference date: 10-11-2022 Through 13-11-2022",

year = "2022",

month = nov,

day = "10",

doi = "10.1145/3574198.3574209",

language = "English",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "68--73",

booktitle = "ICBBE 2022 - Proceeding of 2022 9th International Conference on Biomedical and Bioinformatics Engineering",

}

Qi, W, Du, S, Tang, X & Wang, A 2022, SE-ResNet based vulnerable plaque recognition in IVOCT images. in ICBBE 2022 - Proceeding of 2022 9th International Conference on Biomedical and Bioinformatics Engineering. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 68-73, 9th International Conference on Biomedical and Bioinformatics Engineering, ICBBE 2022, Virtual, Online, Japan, 10/11/22. https://doi.org/10.1145/3574198.3574209

SE-ResNet based vulnerable plaque recognition in IVOCT images. / Qi, Wenliu; Du, Sihui; Tang, Xiaoying et al.
ICBBE 2022 - Proceeding of 2022 9th International Conference on Biomedical and Bioinformatics Engineering. Association for Computing Machinery, 2022. p. 68-73 (ACM International Conference Proceeding Series).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - SE-ResNet based vulnerable plaque recognition in IVOCT images

AU - Qi, Wenliu

AU - Du, Sihui

AU - Tang, Xiaoying

AU - Wang, Ancong

PY - 2022/11/10

Y1 - 2022/11/10

N2 - Acute coronary syndrome (ACS) caused by vulnerable plaques can lead to sudden death. The resolution of intravascular optical coherence tomography (IVOCT) is up to 10 μm, and it has become the first choice for vulnerable plaque recognition. However, it is time-consuming and burdensome for doctors to label vulnerable plaques manually. As a result, it is important to develop an automatic method for vulnerable plaque recognition in IVOCT images. This paper proposes a lightweight and real-time method to identify the main vulnerable plaque areas in IVOCT images. The accuracy rate, recall rate and overlap rate of this method on the test set are 84.8%, 90.1%, and 87.0% respectively, and the recognition quality is 87.2%. The results suggest that our method may assist doctors to recognize vulnerable plaque areas fast and accurately.

AB - Acute coronary syndrome (ACS) caused by vulnerable plaques can lead to sudden death. The resolution of intravascular optical coherence tomography (IVOCT) is up to 10 μm, and it has become the first choice for vulnerable plaque recognition. However, it is time-consuming and burdensome for doctors to label vulnerable plaques manually. As a result, it is important to develop an automatic method for vulnerable plaque recognition in IVOCT images. This paper proposes a lightweight and real-time method to identify the main vulnerable plaque areas in IVOCT images. The accuracy rate, recall rate and overlap rate of this method on the test set are 84.8%, 90.1%, and 87.0% respectively, and the recognition quality is 87.2%. The results suggest that our method may assist doctors to recognize vulnerable plaque areas fast and accurately.

KW - Convolutional neural network

KW - Intravascular optical coherence tomography

KW - Vulnerable plaque

KW - image recognition

UR - http://www.scopus.com/inward/record.url?scp=85150394865&partnerID=8YFLogxK

U2 - 10.1145/3574198.3574209

DO - 10.1145/3574198.3574209

M3 - Conference contribution

AN - SCOPUS:85150394865

T3 - ACM International Conference Proceeding Series

SP - 68

EP - 73

BT - ICBBE 2022 - Proceeding of 2022 9th International Conference on Biomedical and Bioinformatics Engineering

PB - Association for Computing Machinery

T2 - 9th International Conference on Biomedical and Bioinformatics Engineering, ICBBE 2022

Y2 - 10 November 2022 through 13 November 2022

ER -

SE-ResNet based vulnerable plaque recognition in IVOCT images

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this