A multi-model weighted voting-based method for detecting coronary artery disease

Wenxiang Li; Xuezhi Yang; Jing Chen; Xuenan Liu

doi:10.1117/12.2681051

8 June 2023 A multi-model weighted voting-based method for detecting coronary artery disease

Wenxiang Li, Xuezhi Yang, Jing Chen, Xuenan Liu

Proceedings Volume 12707, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023); 127074U (2023) https://doi.org/10.1117/12.2681051
Event: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), 2023, Changsha, China

Abstract

Photoplethysmography (PPG) signal is an effective way for early screening of coronary artery disease (CAD). However, opposite test results of CAD may be drawn from PPG signal ambiguity (e.g., obscure tidal waves and missing repetitive waves). A multi-model weighted voting CAD detection model is proposed in this work to address this problem. Its main processes are as follows. First, the PPG and its second derivative photoplethysmography (SDPPG) features are extracted. Then, feature subsets and training models are obtained by recursive feature elimination algorithms for three base models: support vector machine (SVM), random forest (RF), and logistic regression (LR). The weights of each model are calculated based on the feature subsets and prior knowledge. Finally, each model outputs the classification results according to the weighted voting method. The proposed method evaluates the PPG signals of 130 inpatients and 115 normal subjects collected since, and the accuracy, sensitivity, and specificity of the obtained CAD detection models are 88.64%, 85.32%, and 91.24%. This shows that the multi-model weight voting method can reduce the negative impact of pulse wave uncertainty on a single model for detecting CAD.

Citation Download Citation

Wenxiang Li, Xuezhi Yang, Jing Chen, and Xuenan Liu "A multi-model weighted voting-based method for detecting coronary artery disease", Proc. SPIE 12707, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), 127074U (8 June 2023); https://doi.org/10.1117/12.2681051

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available