Accurately Identifying Coronary Atherosclerotic Heart Disease through Merged Beats of Electrocardiogram
Wang, Xinfeng ; Qi, Mengling ; Dong, Chengzhi ; Zhang, Haoyang LU
; Yang, Yuedong
and Zhao, Huiying
(2022)
2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022
p.1249-1254
- Abstract
Coronary Atherosclerotic Heart Disease (CAHD) is one kind of severe heart disease that is the dominating cause of death from non-communicable diseases worldwide. CAHD can be early detected through pre-symptomatic health check-ups, and the electrocardiogram (ECG) is common for non-invasive health check diagnoses. Traditionally, ECG signals are utilized to extract clinical features that are then input into machine learning methods for training and prediction. While these extracted features are interpretable, they are difficult to break through known features. On the other hand, ECG can be directly input to deep learning techniques, but such methods are usually limited by small sample sizes. Here, we propose to merge multiple beats of raw... (More)
Coronary Atherosclerotic Heart Disease (CAHD) is one kind of severe heart disease that is the dominating cause of death from non-communicable diseases worldwide. CAHD can be early detected through pre-symptomatic health check-ups, and the electrocardiogram (ECG) is common for non-invasive health check diagnoses. Traditionally, ECG signals are utilized to extract clinical features that are then input into machine learning methods for training and prediction. While these extracted features are interpretable, they are difficult to break through known features. On the other hand, ECG can be directly input to deep learning techniques, but such methods are usually limited by small sample sizes. Here, we propose to merge multiple beats of raw signal into one beat, which greatly reduces the complexity while maintaining the raw information. Moreover, we have constructed the largest benchmark dataset for 1113 CAHD patients of 12-lead ECG signals from the UK Biobank database and used the data to train a deep learning model. The results indicated that merged beat signals could achieve the best performance corresponding to an AUC of 0.71 and accuracy of 0.7, which is 4% higher than models using the raw signals and 6% higher than those using the clinical features. Further intuitive interpretation revealed that ST waves in lead II and V3 are the most closely associated with CAHD, consistent with clinical observations.
(Less)
- author
- Wang, Xinfeng
; Qi, Mengling
; Dong, Chengzhi
; Zhang, Haoyang
LU
; Yang, Yuedong
and Zhao, Huiying
- publishing date
- 2022
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- keywords
- Coronary atherosclerotic heart disease, Deep learning, Electrocardiogram, Merged beat, Residual network, XGBoost.
- host publication
- 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022
- editor
- Adjeroh, Donald ; Long, Qi ; Shi, Xinghua ; Guo, Fei ; Hu, Xiaohua ; Aluru, Srinivas ; Narasimhan, Giri ; Wang, Jianxin ; Kang, Mingon ; Mondal, Ananda M. and Liu, Jin
- pages
- 1249 - 1254
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022
- conference location
- Las Vegas, United States
- conference dates
- 2022-12-06 - 2022-12-08
- external identifiers
-
- scopus:85146721439
- ISBN
- 9781665468190
- DOI
- 10.1109/BIBM55620.2022.9995602
- language
- English
- LU publication?
- no
- additional info
- Funding Information: This study has been supported by the National Key R &D Program of China (2020YFB0204803), National Natur al Science Foundation of China (61772566), Guangzhou S &T Research Plan (202007030010). (For the code, supple ment file and dataset of the project, please visit https://git hub.com/andywxf/Predict_CAHD_by_ECG_on_UK_Biobank , send an e-mail to wangxf59@mail2.sysu.edu.cn.) We'd like to acknowledge the UK Biobank to make the data available.UK Biobank data ID/protocol number 65805. Publisher Copyright: © 2022 IEEE.
- id
- af6154fc-d225-4ea9-a362-1d2574f13d83
- date added to LUP
- 2024-02-05 15:10:25
- date last changed
- 2024-02-06 08:18:15
@inproceedings{af6154fc-d225-4ea9-a362-1d2574f13d83,
abstract = {{<p>Coronary Atherosclerotic Heart Disease (CAHD) is one kind of severe heart disease that is the dominating cause of death from non-communicable diseases worldwide. CAHD can be early detected through pre-symptomatic health check-ups, and the electrocardiogram (ECG) is common for non-invasive health check diagnoses. Traditionally, ECG signals are utilized to extract clinical features that are then input into machine learning methods for training and prediction. While these extracted features are interpretable, they are difficult to break through known features. On the other hand, ECG can be directly input to deep learning techniques, but such methods are usually limited by small sample sizes. Here, we propose to merge multiple beats of raw signal into one beat, which greatly reduces the complexity while maintaining the raw information. Moreover, we have constructed the largest benchmark dataset for 1113 CAHD patients of 12-lead ECG signals from the UK Biobank database and used the data to train a deep learning model. The results indicated that merged beat signals could achieve the best performance corresponding to an AUC of 0.71 and accuracy of 0.7, which is 4% higher than models using the raw signals and 6% higher than those using the clinical features. Further intuitive interpretation revealed that ST waves in lead II and V3 are the most closely associated with CAHD, consistent with clinical observations.</p>}},
author = {{Wang, Xinfeng and Qi, Mengling and Dong, Chengzhi and Zhang, Haoyang and Yang, Yuedong and Zhao, Huiying}},
booktitle = {{2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022}},
editor = {{Adjeroh, Donald and Long, Qi and Shi, Xinghua and Guo, Fei and Hu, Xiaohua and Aluru, Srinivas and Narasimhan, Giri and Wang, Jianxin and Kang, Mingon and Mondal, Ananda M. and Liu, Jin}},
isbn = {{9781665468190}},
keywords = {{Coronary atherosclerotic heart disease; Deep learning; Electrocardiogram; Merged beat; Residual network; XGBoost.}},
language = {{eng}},
pages = {{1249--1254}},
publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
title = {{Accurately Identifying Coronary Atherosclerotic Heart Disease through Merged Beats of Electrocardiogram}},
url = {{http://dx.doi.org/10.1109/BIBM55620.2022.9995602}},
doi = {{10.1109/BIBM55620.2022.9995602}},
year = {{2022}},
}