Reduction of Spike-like Noise in Clinical Practice for Thoracic Electrical Impedance Tomography Using Robust Principal Component Analysis

Thoracic electrical impedance tomography (EIT) provides real-time, bedside imaging of pulmonary function and has demonstrated significant clinical value in guiding treatment strategies for critically ill patients. However, the practical application of EIT remains challenging due to its susceptibilit...

Full description

Saved in:
Bibliographic Details
Main Authors: Meng Dai, Xiaopeng Li, Zhanqi Zhao, Lin Yang
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Bioengineering
Subjects:
Online Access:https://www.mdpi.com/2306-5354/12/4/402
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850144654986575872
author Meng Dai
Xiaopeng Li
Zhanqi Zhao
Lin Yang
author_facet Meng Dai
Xiaopeng Li
Zhanqi Zhao
Lin Yang
author_sort Meng Dai
collection DOAJ
description Thoracic electrical impedance tomography (EIT) provides real-time, bedside imaging of pulmonary function and has demonstrated significant clinical value in guiding treatment strategies for critically ill patients. However, the practical application of EIT remains challenging due to its susceptibility to measurement disturbances, such as electrode contact problems and patient movement. These disturbances often manifest as spike-like noise that can severely degrade EIT image quality. To address this issue, we propose a robust Principal Component Analysis (RPCA)-based approach that models EIT data as the sum of a low-rank matrix and a sparse matrix. The low-rank matrix captures the underlying physiological signals, while the sparse matrix contains spike-like noise components. In simulation studies considering different spike magnitudes, widths and channels, all the image correlation coefficients between RPCA-processed images and the ground truth exceeded 0.99, and the image error of the original fEIT image with spike-like noise was much larger than that after RPCA processing. In eight patient cases, RPCA significantly improved the image quality (image error: <i>p</i> < 0.001; image correlation coefficient: <i>p</i> < 0.001) and enhanced the clinical EIT-based indexes accuracy (<i>p</i> < 0.001). Therefore, we conclude that RPCA is a promising technique for reducing spike-like noise in clinical EIT data, thereby improving data quality and potentially facilitating broader clinical application of EIT.
format Article
id doaj-art-93880293c6ab4e9b9137c8341ce0e22d
institution OA Journals
issn 2306-5354
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Bioengineering
spelling doaj-art-93880293c6ab4e9b9137c8341ce0e22d2025-08-20T02:28:18ZengMDPI AGBioengineering2306-53542025-04-0112440210.3390/bioengineering12040402Reduction of Spike-like Noise in Clinical Practice for Thoracic Electrical Impedance Tomography Using Robust Principal Component AnalysisMeng Dai0Xiaopeng Li1Zhanqi Zhao2Lin Yang3Department of Biomedical Engineering, The Fourth Military Medical University, Xi’an 710032, ChinaState Key Laboratory of Radio Frequency Heterogeneous Integration, Shenzhen University, Shenzhen 518060, ChinaSchool of Biomedical Engineering, Guangzhou Medical University, Guangzhou 511436, ChinaDepartment of Aerospace Medicine, The Fourth Military Medical University, Xi’an 710032, ChinaThoracic electrical impedance tomography (EIT) provides real-time, bedside imaging of pulmonary function and has demonstrated significant clinical value in guiding treatment strategies for critically ill patients. However, the practical application of EIT remains challenging due to its susceptibility to measurement disturbances, such as electrode contact problems and patient movement. These disturbances often manifest as spike-like noise that can severely degrade EIT image quality. To address this issue, we propose a robust Principal Component Analysis (RPCA)-based approach that models EIT data as the sum of a low-rank matrix and a sparse matrix. The low-rank matrix captures the underlying physiological signals, while the sparse matrix contains spike-like noise components. In simulation studies considering different spike magnitudes, widths and channels, all the image correlation coefficients between RPCA-processed images and the ground truth exceeded 0.99, and the image error of the original fEIT image with spike-like noise was much larger than that after RPCA processing. In eight patient cases, RPCA significantly improved the image quality (image error: <i>p</i> < 0.001; image correlation coefficient: <i>p</i> < 0.001) and enhanced the clinical EIT-based indexes accuracy (<i>p</i> < 0.001). Therefore, we conclude that RPCA is a promising technique for reducing spike-like noise in clinical EIT data, thereby improving data quality and potentially facilitating broader clinical application of EIT.https://www.mdpi.com/2306-5354/12/4/402electrical impedance tomographyspike-like noiserobust principal component analysis
spellingShingle Meng Dai
Xiaopeng Li
Zhanqi Zhao
Lin Yang
Reduction of Spike-like Noise in Clinical Practice for Thoracic Electrical Impedance Tomography Using Robust Principal Component Analysis
Bioengineering
electrical impedance tomography
spike-like noise
robust principal component analysis
title Reduction of Spike-like Noise in Clinical Practice for Thoracic Electrical Impedance Tomography Using Robust Principal Component Analysis
title_full Reduction of Spike-like Noise in Clinical Practice for Thoracic Electrical Impedance Tomography Using Robust Principal Component Analysis
title_fullStr Reduction of Spike-like Noise in Clinical Practice for Thoracic Electrical Impedance Tomography Using Robust Principal Component Analysis
title_full_unstemmed Reduction of Spike-like Noise in Clinical Practice for Thoracic Electrical Impedance Tomography Using Robust Principal Component Analysis
title_short Reduction of Spike-like Noise in Clinical Practice for Thoracic Electrical Impedance Tomography Using Robust Principal Component Analysis
title_sort reduction of spike like noise in clinical practice for thoracic electrical impedance tomography using robust principal component analysis
topic electrical impedance tomography
spike-like noise
robust principal component analysis
url https://www.mdpi.com/2306-5354/12/4/402
work_keys_str_mv AT mengdai reductionofspikelikenoiseinclinicalpracticeforthoracicelectricalimpedancetomographyusingrobustprincipalcomponentanalysis
AT xiaopengli reductionofspikelikenoiseinclinicalpracticeforthoracicelectricalimpedancetomographyusingrobustprincipalcomponentanalysis
AT zhanqizhao reductionofspikelikenoiseinclinicalpracticeforthoracicelectricalimpedancetomographyusingrobustprincipalcomponentanalysis
AT linyang reductionofspikelikenoiseinclinicalpracticeforthoracicelectricalimpedancetomographyusingrobustprincipalcomponentanalysis