Automatic atrial fibrillation detection based on feature engineering and deep learning

Chi Ma; Xiaoxue Li; Xin Yang; Baoming Pu; Jianzhong Qiao

doi:10.1117/12.3055335

12 December 2024 Automatic atrial fibrillation detection based on feature engineering and deep learning

Chi Ma, Xiaoxue Li, Xin Yang, Baoming Pu, Jianzhong Qiao

Author Affiliations +

Proceedings Volume 13439, Fourth International Conference on Testing Technology and Automation Engineering (TTAE 2024); 134391L (2024) https://doi.org/10.1117/12.3055335
Event: Fourth International Conference on Testing Technology and Automation Engineering (TTAE 2024), 2024, Xiamen, China

Abstract

As cardiovascular disease rates continue to climb, atrial fibrillation (AF), a common type of arrhythmia, increasingly threatens public health. Traditional electrocardiogram (ECG) diagnostics depend heavily on the subjective interpretation by doctors, making them prone to human error. This paper presents a method that combines traditional feature engineering with deep learning techniques. It extracts key statistical features of the ECG signal through feature engineering, while also using a deep learning model based on a shifted window attention mechanism to further extract temporal dynamic features. Experimental results demonstrate that this method surpasses traditional techniques in detecting AF, especially improving specificity and sensitivity, thereby significantly enhancing diagnostic accuracy.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Chi Ma, Xiaoxue Li, Xin Yang, Baoming Pu, and Jianzhong Qiao "Automatic atrial fibrillation detection based on feature engineering and deep learning", Proc. SPIE 13439, Fourth International Conference on Testing Technology and Automation Engineering (TTAE 2024), 134391L (12 December 2024); https://doi.org/10.1117/12.3055335

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