Gated Recurrent Unit-based Heart Sound Analysis for Heart Failure Screening

Overview

Journal Biomed Eng Online

Publisher Biomed Central

Specialty Biomedical Engineering

Date 2020 Jan 15

PMID 31931811

Citations 13

Authors

Shan Gao

Yineng Zheng

Xingming Guo

Affiliations

Soon will be listed here.

Abstract

Background: Heart failure (HF) is a type of cardiovascular disease caused by abnormal cardiac structure and function. Early screening of HF has important implication for treatment in a timely manner. Heart sound (HS) conveys relevant information related to HF; this study is therefore based on the analysis of HS signals. The objective is to develop an efficient tool to identify subjects of normal, HF with preserved ejection fraction and HF with reduced ejection fraction automatically.

Methods: We proposed a novel HF screening framework based on gated recurrent unit (GRU) model in this study. The logistic regression-based hidden semi-Markov model was adopted to segment HS frames. Normalized frames were taken as the input of the proposed model which can automatically learn the deep features and complete the HF screening without de-nosing and hand-crafted feature extraction.

Results: To evaluate the performance of proposed model, three methods are used for comparison. The results show that the GRU model gives a satisfactory performance with average accuracy of 98.82%, which is better than other comparison models.

Conclusion: The proposed GRU model can learn features from HS directly, which means it can be independent of expert knowledge. In addition, the good performance demonstrates the effectiveness of HS analysis for HF early screening.

Citing Articles

Optimizing extubation success: a comparative analysis of time series algorithms and activation functions.

Huang K, Lin C, Chi S, Hsu Y, Xu J Front Comput Neurosci. 2024; 18:1456771.

PMID: 39429247 PMC: 11486667. DOI: 10.3389/fncom.2024.1456771.

Deep Learning in Heart Sound Analysis: From Techniques to Clinical Applications.

Zhao Q, Geng S, Wang B, Sun Y, Nie W, Bai B Health Data Sci. 2024; 4:0182.

PMID: 39387057 PMC: 11461928. DOI: 10.34133/hds.0182.

Applications of deep learning models in precision prediction of survival rates for heart failure patients.

Zhang Q, Xu D Technol Health Care. 2024; 32(S1):329-337.

PMID: 38759059 PMC: 11191484. DOI: 10.3233/THC-248029.

[Research on bark-frequency spectral coefficients heart sound classification algorithm based on multiple window time-frequency reassignment].

Xia J, Sun J, Yang H, Pan J, Guo T, Wang W Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2024; 41(1):51-59.

PMID: 38403604 PMC: 10894746. DOI: 10.7507/1001-5515.202212037.

Association between phonocardiography and echocardiography in heart failure patients with preserved ejection fraction.

Luo H, Weerts J, Bekkers A, Achten A, Lievens S, Smeets K Eur Heart J Digit Health. 2023; 4(1):4-11.

PMID: 36743874 PMC: 9890082. DOI: 10.1093/ehjdh/ztac073.

References

Mabote T, Wong K, Cleland J . The utility of novel non-invasive technologies for remote hemodynamic monitoring in chronic heart failure. Expert Rev Cardiovasc Ther. 2014; 12(8):923-8. DOI: 10.1586/14779072.2014.935339. View

Savarese G, Orsini N, Hage C, Vedin O, Cosentino F, Rosano G . Utilizing NT-proBNP for Eligibility and Enrichment in Trials in HFpEF, HFmrEF, and HFrEF. JACC Heart Fail. 2018; 6(3):246-256. DOI: 10.1016/j.jchf.2017.12.014. View

Yu R, Zheng Y, Zhang R, Jiang Y, Poon C . Using a Multi-Task Recurrent Neural Network With Attention Mechanisms to Predict Hospital Mortality of Patients. IEEE J Biomed Health Inform. 2019; 24(2):486-492. DOI: 10.1109/JBHI.2019.2916667. View

Zheng Y, Guo X, Qin J, Xiao S . Computer-assisted diagnosis for chronic heart failure by the analysis of their cardiac reserve and heart sound characteristics. Comput Methods Programs Biomed. 2015; 122(3):372-83. DOI: 10.1016/j.cmpb.2015.09.001. View

Hochreiter S, Schmidhuber J . Long short-term memory. Neural Comput. 1997; 9(8):1735-80. DOI: 10.1162/neco.1997.9.8.1735. View

Yildirim O, Plawiak P, Tan R, Rajendra Acharya U . Arrhythmia detection using deep convolutional neural network with long duration ECG signals. Comput Biol Med. 2018; 102:411-420. DOI: 10.1016/j.compbiomed.2018.09.009. View

Xanthopoulos A, Triposkiadis F, Starling R . Heart failure with preserved ejection fraction: Classification based upon phenotype is essential for diagnosis and treatment. Trends Cardiovasc Med. 2018; 28(6):392-400. DOI: 10.1016/j.tcm.2018.01.001. View

Xu L, Huang X, Ma J, Huang J, Fan Y, Li H . Value of three-dimensional strain parameters for predicting left ventricular remodeling after ST-elevation myocardial infarction. Int J Cardiovasc Imaging. 2017; 33(5):663-673. DOI: 10.1007/s10554-016-1053-3. View

Gao Z, Chung J, Abdelrazek M, Leung S, Hau W, Xian Z . Privileged Modality Distillation for Vessel Border Detection in Intracoronary Imaging. IEEE Trans Med Imaging. 2019; 39(5):1524-1534. DOI: 10.1109/TMI.2019.2952939. View

10.

Xu C, Xu L, Gao Z, Zhao S, Zhang H, Zhang Y . Direct delineation of myocardial infarction without contrast agents using a joint motion feature learning architecture. Med Image Anal. 2018; 50:82-94. DOI: 10.1016/j.media.2018.09.001. View

11.

Michielli N, Rajendra Acharya U, Molinari F . Cascaded LSTM recurrent neural network for automated sleep stage classification using single-channel EEG signals. Comput Biol Med. 2019; 106:71-81. DOI: 10.1016/j.compbiomed.2019.01.013. View

12.

Gao Z, Xiong H, Liu X, Zhang H, Ghista D, Wu W . Robust estimation of carotid artery wall motion using the elasticity-based state-space approach. Med Image Anal. 2017; 37:1-21. DOI: 10.1016/j.media.2017.01.004. View

13.

Beritelli F, Capizzi G, Lo Sciuto G, Napoli C, Scaglione F . Automatic heart activity diagnosis based on Gram polynomials and probabilistic neural networks. Biomed Eng Lett. 2019; 8(1):77-85. PMC: 6208560. DOI: 10.1007/s13534-017-0046-z. View

14.

Zheng Y, Guo X . Identification of chronic heart failure using linear and nonlinear analysis of heart sound. Annu Int Conf IEEE Eng Med Biol Soc. 2017; 2017:4586-4589. DOI: 10.1109/EMBC.2017.8037877. View

15.

Faxen U, Hage C, Benson L, Zabarovskaja S, Andreasson A, Donal E . HFpEF and HFrEF Display Different Phenotypes as Assessed by IGF-1 and IGFBP-1. J Card Fail. 2016; 23(4):293-303. DOI: 10.1016/j.cardfail.2016.06.008. View

16.

Liu Y, Guo X, Zheng Y . An Automatic Approach Using ELM Classifier for HFpEF Identification Based on Heart Sound Characteristics. J Med Syst. 2019; 43(9):285. DOI: 10.1007/s10916-019-1415-1. View

17.

Giordano N, Knaflitz M . A Novel Method for Measuring the Timing of Heart Sound Components through Digital Phonocardiography. Sensors (Basel). 2019; 19(8). PMC: 6515005. DOI: 10.3390/s19081868. View

18.

Gao Z, Li Y, Sun Y, Yang J, Xiong H, Zhang H . Motion Tracking of the Carotid Artery Wall From Ultrasound Image Sequences: a Nonlinear State-Space Approach. IEEE Trans Med Imaging. 2017; 37(1):273-283. DOI: 10.1109/TMI.2017.2746879. View

19.

Gao Z, Wu S, Liu Z, Luo J, Zhang H, Gong M . Learning the implicit strain reconstruction in ultrasound elastography using privileged information. Med Image Anal. 2019; 58:101534. DOI: 10.1016/j.media.2019.101534. View

20.

Nair N, Gupta S, Collier I, Gongora E, Vijayaraghavan K . Can microRNAs emerge as biomarkers in distinguishing HFpEF versus HFrEF?. Int J Cardiol. 2014; 175(3):395-9. DOI: 10.1016/j.ijcard.2014.06.027. View