CECGSR: Circular ECG Super-Resolution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Honggui, Zhang, Zhengyang, Li, Dingtai, Chen, Sinan, Hossain, Nahid Md Lokman, Lu, Hantao, Wang, Ruobing, Xu, Xinfeng, Qin, Yinlu, Feng, Yuting, Trocan, Maria, Galayko, Dimitri, Amara, Amara, Sawan, Mohamad
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908869603622912
author Li, Honggui
Zhang, Zhengyang
Li, Dingtai
Chen, Sinan
Hossain, Nahid Md Lokman
Lu, Hantao
Wang, Ruobing
Xu, Xinfeng
Qin, Yinlu
Feng, Yuting
Trocan, Maria
Galayko, Dimitri
Amara, Amara
Sawan, Mohamad
author_facet Li, Honggui
Zhang, Zhengyang
Li, Dingtai
Chen, Sinan
Hossain, Nahid Md Lokman
Lu, Hantao
Wang, Ruobing
Xu, Xinfeng
Qin, Yinlu
Feng, Yuting
Trocan, Maria
Galayko, Dimitri
Amara, Amara
Sawan, Mohamad
contents Background and Objective: The electrocardiogram (ECG) plays a crucial role in the diagnosis and treatment of various cardiac diseases. ECG signals suffer from low-resolution (LR) due to the use of convenient acquisition devices, as well as internal and external noises and artifacts. Classical ECG super-resolution (ECGSR) methods adopt an open-loop architecture that converts LR ECG signals to super-resolution (SR) ones. According to the theory of automatic control, a closed-loop framework exhibits superior dynamic and static performance compared with its open-loop counterpart. Methods: This paper proposes a closed-loop approach, termed circular ECGSR (CECGSR), which models the degradation process from SR ECG signals to LR ones. The negative feedback mechanism of the closed-loop system is based on the differences between the LR ECG signals. A mathematical loop equation is constructed to characterize the closed-loop infrastructure. The Taylor series expansion is employed to demonstrate the near-zero steady-state error of the proposed method. A Plug-and-Play strategy is considered to establish the SR unit of the proposed architecture, leveraging any existing advanced open-loop ECGSR methods. This paper also presents Transformer model based open-loop ECGSR and closed-loop CECGSR algorithms. Results: Simulation experiments on both noiseless and noisy subsets of the Physikalisch-Technische Bundesanstalt-Extra Large (PTB-XL) datasets demonstrate that the proposed CECGSR outperforms state-of-the-art open-loop ECGSR algorithms in the reconstruction performance of ECG signals. Conclusions: The proposed method will efficiently enrich ECG signal details and remove ECG signal artifacts in clinical applications.
format Preprint
id arxiv_https___arxiv_org_abs_2508_11658
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CECGSR: Circular ECG Super-Resolution
Li, Honggui
Zhang, Zhengyang
Li, Dingtai
Chen, Sinan
Hossain, Nahid Md Lokman
Lu, Hantao
Wang, Ruobing
Xu, Xinfeng
Qin, Yinlu
Feng, Yuting
Trocan, Maria
Galayko, Dimitri
Amara, Amara
Sawan, Mohamad
Signal Processing
Background and Objective: The electrocardiogram (ECG) plays a crucial role in the diagnosis and treatment of various cardiac diseases. ECG signals suffer from low-resolution (LR) due to the use of convenient acquisition devices, as well as internal and external noises and artifacts. Classical ECG super-resolution (ECGSR) methods adopt an open-loop architecture that converts LR ECG signals to super-resolution (SR) ones. According to the theory of automatic control, a closed-loop framework exhibits superior dynamic and static performance compared with its open-loop counterpart. Methods: This paper proposes a closed-loop approach, termed circular ECGSR (CECGSR), which models the degradation process from SR ECG signals to LR ones. The negative feedback mechanism of the closed-loop system is based on the differences between the LR ECG signals. A mathematical loop equation is constructed to characterize the closed-loop infrastructure. The Taylor series expansion is employed to demonstrate the near-zero steady-state error of the proposed method. A Plug-and-Play strategy is considered to establish the SR unit of the proposed architecture, leveraging any existing advanced open-loop ECGSR methods. This paper also presents Transformer model based open-loop ECGSR and closed-loop CECGSR algorithms. Results: Simulation experiments on both noiseless and noisy subsets of the Physikalisch-Technische Bundesanstalt-Extra Large (PTB-XL) datasets demonstrate that the proposed CECGSR outperforms state-of-the-art open-loop ECGSR algorithms in the reconstruction performance of ECG signals. Conclusions: The proposed method will efficiently enrich ECG signal details and remove ECG signal artifacts in clinical applications.
title CECGSR: Circular ECG Super-Resolution
topic Signal Processing
url https://arxiv.org/abs/2508.11658