1
|
Yang M, Liu W, Zhang H. A robust multiple heartbeats classification with weight-based loss based on convolutional neural network and bidirectional long short-term memory. Front Physiol 2022; 13:982537. [PMID: 36545286 PMCID: PMC9760867 DOI: 10.3389/fphys.2022.982537] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/30/2022] [Accepted: 11/18/2022] [Indexed: 12/09/2022] Open
Abstract
Background: Analysis of electrocardiogram (ECG) provides a straightforward and non-invasive approach for cardiologists to diagnose and classify the nature and severity of variant cardiac diseases including cardiac arrhythmia. However, the interpretation and analysis of ECG are highly working-load demanding, and the subjective may lead to false diagnoses and heartbeats classification. In recent years, many deep learning works showed an excellent role in accurate heartbeats classification. However, the imbalance of heartbeat classes is universal in most of the available ECG databases since abnormal heartbeats are always relatively rare in real life scenarios. In addition, many existing approaches achieved prominent results by removing noise and extracting features in data preprocessing, which relies heavily on powerful computers. It is a pressing need to develop efficient and automatic light weighted algorithms for accurate heartbeats classification that can be used in portable ECG sensors. Objective: This study aims at developing a robust and efficient deep learning method, which can be embedded into wearable or portable ECG monitors for classifying heartbeats. Methods: We proposed a novel and light weighted deep learning architecture with weight-based loss based on a convolutional neural network (CNN) and bidirectional long short-term memory (Bi-LSTM) that can automatically identify five types of ECG heartbeats according to the AAMI EC57 standard. It was also true that the raw ECG signals were simply segmented without noise removal and other feature extraction processing. Moreover, to tackle the challenge of classification bias due to imbalanced ECG datasets for different types of arrhythmias, we introduced a weight-based loss function to reduce the influence of over-weighted categories in the ECG dataset. For avoiding the influence of the division of validation dataset, k-fold method was adopted to improve the reliability of the model. Results: The proposed algorithm is trained and tested on MIT-BIH Arrhythmia Database, and achieves an average of 99.33% accuracy, 93.67% sensitivity, 99.18% specificity, 89.85% positive prediction, and 91.65% F1 score.
Collapse
Affiliation(s)
- Mengting Yang
- Key Laboratory of Medical Electrophysiology, Ministry of Education and Medical Electrophysiological Key Laboratory of Sichuan Province, (Collaborative Innovation Center for Prevention of Cardiovascular Diseases), Institute of Cardiovascular Research, Southwest Medical University, Luzhou, China,School of Medical Information and Engineering, Southwest Medical University, Luzhou, China,School of Biomedical Engineering and Instrument Science, Zhejiang University, Hangzhou, China
| | - Weichao Liu
- Key Laboratory of Medical Electrophysiology, Ministry of Education and Medical Electrophysiological Key Laboratory of Sichuan Province, (Collaborative Innovation Center for Prevention of Cardiovascular Diseases), Institute of Cardiovascular Research, Southwest Medical University, Luzhou, China
| | - Henggui Zhang
- Key Laboratory of Medical Electrophysiology, Ministry of Education and Medical Electrophysiological Key Laboratory of Sichuan Province, (Collaborative Innovation Center for Prevention of Cardiovascular Diseases), Institute of Cardiovascular Research, Southwest Medical University, Luzhou, China,Department of Physics and Astronomy, The University of Manchester, Manchester, United Kingdom,*Correspondence: Henggui Zhang,
| |
Collapse
|
2
|
Cai Z, Cheng H, Xing Y, Chen F, Zhang Y, Cui C. Autonomic nervous activity analysis based on visibility graph complex networks and skin sympathetic nerve activity. Front Physiol 2022; 13:1001415. [PMID: 36160855 PMCID: PMC9500413 DOI: 10.3389/fphys.2022.1001415] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/23/2022] [Accepted: 08/17/2022] [Indexed: 11/19/2022] Open
Abstract
Background: Autonomic nerve system (ANS) plays an important role in regulating cardiovascular function and cerebrovascular function. Traditional heart rate variation (HRV) and emerging skin sympathetic nerve activity (SKNA) analyses from ultra-short-time (UST) data cannot fully reveal neural activity, thereby quantitatively reflect ANS intensity. Methods: Electrocardiogram and SKNA from sixteen patients (seven cerebral hemorrhage (CH) patients and nine control group (CO) patients) were recorded using a portable device. Ten derived HRV (mean, standard deviation and root mean square difference of sinus RR intervals (NNmean, SDNN and RMSSD), ultra-low frequency (<0.003 Hz, uLF), very low frequency ([0.003 Hz, 0.04 Hz), vLF), low frequency ([0.04 Hz, 0.15 Hz), LF) and high frequency power ([0.15 Hz, 0.4 Hz), HF), ratio of LF to HF (LF/HF), the standard deviation of instantaneous beat-to-beat R-R interval variability (SD1), and approximate entropy (ApEn)) and ten visibility graph (VG) features (diameter (Dia), average node degree (aND), average shortest-path length (aSPL), clustering coefficient (CC), average closeness centrality (aCC), transitivity (Trans), average degree centrality (aDC), link density (LD), sMetric (sM) and graph energy (GE) of the constructed complex network) were compared on 5-min and UST segments to verify their validity and robustness in discriminating CH and CO under different data lengths. Besides, their potential for quantifying ANS-Load were also investigated. Results: The validation results of HRV and VG features in discriminating CH from CO showed that VG features were more clearly distinguishable between the two groups than HRV features. For effectiveness evaluation of analyzing ANS on UST segment, the NNmean, SDNN, RMSSD, LF, HF and LF/HF in HRV features and the CC, Trans, Dia and GE of VG features remained stable in both activated and inactivated segments across all data lengths. The capability of HRV and VG features in quantifying ANS-Load were evaluated and compared under different ANS-Load, the results showed that most HRV features (SDNN, LFHF, RMSSD, vLF, LF and HF) and almost all VG features were correlated to sympathetic nerve activity intensity. Conclusions: The proposed autonomic nervous activity analysis method based on VG and SKNA offers a new insight into ANS assessment in UST segments and ANS-Load quantification.
Collapse
Affiliation(s)
- Zhipeng Cai
- School of Instrument Science and Engineering, Southeast University, Nanjing, China
- *Correspondence: Zhipeng Cai, ; Chang Cui,
| | - Hongyi Cheng
- Department of Cardiology, Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
- Gusu School, Nanjing Medical University, The Affiliated Suzhou Hospital of Nanjing Medical University, Suzhou, China
| | - Yantao Xing
- School of Instrument Science and Engineering, Southeast University, Nanjing, China
| | - Feifei Chen
- School of Instrument Science and Engineering, Southeast University, Nanjing, China
| | - Yike Zhang
- Department of Cardiology, Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
| | - Chang Cui
- Department of Cardiology, Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
- *Correspondence: Zhipeng Cai, ; Chang Cui,
| |
Collapse
|
3
|
Fuadah YN, Lim KM. Optimal Classification of Atrial Fibrillation and Congestive Heart Failure Using Machine Learning. Front Physiol 2022; 12:761013. [PMID: 35185594 PMCID: PMC8850703 DOI: 10.3389/fphys.2021.761013] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/19/2021] [Accepted: 12/29/2021] [Indexed: 11/13/2022] Open
Abstract
Cardiovascular disorders, including atrial fibrillation (AF) and congestive heart failure (CHF), are the significant causes of mortality worldwide. The diagnosis of cardiovascular disorders is heavily reliant on ECG signals. Therefore, extracting significant features from ECG signals is the most challenging aspect of representing each condition of ECG signal. Earlier studies have claimed that the Hjorth descriptor is assigned as a simple feature extraction algorithm capable of class separation among AF, CHF, and normal sinus rhythm (NSR) conditions. However, due to noise interference, certain features do not represent the characteristics of the ECG signals. This study addressed this critical gap by applying the discrete wavelet transform (DWT) to decompose the ECG signals into sub-bands and extracting Hjorth descriptor features and entropy-based features in the DWT domain. Therefore, the calculation of Hjorth descriptor and entropy-based features performed on each sub-band will produce more detailed information of ECG signals. The optimization of various classifier algorithms, including k-nearest neighbor (k-NN), support vector machine (SVM), random forest (RF), artificial neural network (ANN), and radial basis function network (RBFN), was investigated to provide the best system performance. This study obtained an accuracy of 100% for the k-NN, SVM, RF, and ANN classifiers, respectively, and 97% for the RBFN classifier. The results demonstrated that the optimization of the classifier algorithm could improve the classification accuracy of AF, CHF, and NSR conditions, compared to earlier studies.
Collapse
Affiliation(s)
- Yunendah Nur Fuadah
- Computationa Medicine Lab, Department of IT Convergence Engineering, Kumoh National Institute of Technology, Gumi, South Korea
| | - Ki Moo Lim
- Computational Medicine Lab, Department of Medical IT Convergence Engineering, Kumoh National Institute of Technology, Gumi, South Korea
- *Correspondence: Ki Moo Lim,
| |
Collapse
|
4
|
Rahul J, Sharma LD, Bohat VK. Short duration Vectorcardiogram based inferior myocardial infarction detection: class and subject-oriented approach. BIOMED ENG-BIOMED TE 2021; 66:489-501. [PMID: 33939896 DOI: 10.1515/bmt-2020-0329] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/02/2020] [Accepted: 03/30/2021] [Indexed: 12/17/2022]
Abstract
Myocardial infarction (MI) happens when blood stops circulating to an explicit segment of the heart causing harm to the heart muscles. Vectorcardiography (VCG) is a technique of recording direction and magnitude of the signals that are produced by the heart in a 3-lead representation. In this work, we present a technique for detection of MI in the inferior portion of heart using short duration VCG signals. The raw signal was pre-processed using the median and Savitzky-Golay (SG) filter. The Stationary Wavelet Transform (SWT) was used for time-invariant decomposition of the signal followed by feature extraction. The selected features using minimum-redundancy-maximum-relevance (mRMR) based feature selection method were applied to the supervised classification methods. The efficacy of the proposed method was assessed under both class-oriented and a more real-life subject-oriented approach. An accuracy of 99.14 and 89.37% were achieved respectively. Results of the proposed technique are better than existing state-of-art methods and used VCG segment is shorter. Thus, a shorter segment and a high accuracy can be helpful in the automation of timely and reliable detection of MI. The satisfactory performance achieved in the subject-oriented approach shows reliability and applicability of the proposed technique.
Collapse
Affiliation(s)
- Jagdeep Rahul
- Department of Electronics & Communication Engineering, Rajiv Gandhi University, Itanagar, Arunachal Pradesh, India
| | - Lakhan Dev Sharma
- School of Electronics Engineering, VIT-AP University, Amaravati, Andhra Pradesh, India
| | - Vijay Kumar Bohat
- Department of Computer Science & Engineering, Bennett University, Greater Noida, Uttar Pradesh, India
| |
Collapse
|
5
|
Gan Y, Shi JC, He WM, Sun FJ. Parallel classification model of arrhythmia based on DenseNet-BiLSTM. Biocybern Biomed Eng 2021. [DOI: 10.1016/j.bbe.2021.09.001] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/30/2022]
|
6
|
Nejadeh M, Bayat P, Kheirkhah J, Moladoust H. Predicting the response to cardiac resynchronization therapy (CRT) using the deep learning approach. Biocybern Biomed Eng 2021. [DOI: 10.1016/j.bbe.2021.05.002] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/03/2023]
|
7
|
Faruk N, Abdulkarim A, Emmanuel I, Folawiyo YY, Adewole KS, Mojeed HA, Oloyede AA, Olawoyin LA, Sikiru IA, Nehemiah M, Ya'u Gital A, Chiroma H, Ogunmodede JA, Almutairi M, Katibi IA. A comprehensive survey on low-cost ECG acquisition systems: Advances on design specifications, challenges and future direction. Biocybern Biomed Eng 2021. [DOI: 10.1016/j.bbe.2021.02.007] [Citation(s) in RCA: 7] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/05/2023]
|
8
|
Lei M, Li J, Li M, Zou L, Yu H. An Improved UNet++ Model for Congestive Heart Failure Diagnosis Using Short-Term RR Intervals. Diagnostics (Basel) 2021; 11:diagnostics11030534. [PMID: 33809773 PMCID: PMC8002263 DOI: 10.3390/diagnostics11030534] [Citation(s) in RCA: 10] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/21/2021] [Revised: 03/13/2021] [Accepted: 03/14/2021] [Indexed: 12/19/2022] Open
Abstract
Congestive heart failure (CHF), a progressive and complex syndrome caused by ventricular dysfunction, is difficult to detect at an early stage. Heart rate variability (HRV) was proposed as a prognostic indicator for CHF. Inspired by the success of 2-D UNet++ in medical image segmentation, in this paper, we introduce an end-to-end encoder-decoder model to detect CHF using HRV signals. The developed model enhances the UNet++ model with Squeeze-and-Excitation (SE) residual blocks to extract deep features hierarchically and distinguish CHF patients from normal subjects. Two open-source databases are utilized for evaluating the proposed method, and three segment lengths of intervals between successive R-peaks are employed in comparison with state-of-the-art methods. The proposed method achieves an accuracy of 85.64%, 86.65% and 88.79% when 500, 1000 and 2000 RR intervals are utilized, respectively. It demonstrates that HRV evaluation based on deep learning can be an important tool for early detection of CHF, and may assist clinicians in achieving timely and accurate diagnoses.
Collapse
Affiliation(s)
- Meng Lei
- School of Information and Electrical Control Engineering, China University of Mining and Technology, Xuzhou 221116, China; (M.L.); (J.L.); (M.L.)
| | - Jia Li
- School of Information and Electrical Control Engineering, China University of Mining and Technology, Xuzhou 221116, China; (M.L.); (J.L.); (M.L.)
| | - Ming Li
- School of Information and Electrical Control Engineering, China University of Mining and Technology, Xuzhou 221116, China; (M.L.); (J.L.); (M.L.)
| | - Liang Zou
- School of Information and Electrical Control Engineering, China University of Mining and Technology, Xuzhou 221116, China; (M.L.); (J.L.); (M.L.)
- Correspondence:
| | - Han Yu
- School of Computer Science and Engineering (SCSE), Nanyang Technological University (NTU), Singapore 639798, Singapore;
| |
Collapse
|
9
|
A Novel Approach for Coronary Artery Disease Diagnosis using Hybrid Particle Swarm Optimization based Emotional Neural Network. Biocybern Biomed Eng 2020. [DOI: 10.1016/j.bbe.2020.09.005] [Citation(s) in RCA: 13] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/19/2022]
|
10
|
Cherian RP, Thomas N, Venkitachalam S. Weight optimized neural network for heart disease prediction using hybrid lion plus particle swarm algorithm. J Biomed Inform 2020; 110:103543. [PMID: 32858167 DOI: 10.1016/j.jbi.2020.103543] [Citation(s) in RCA: 13] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/12/2019] [Revised: 08/01/2020] [Accepted: 08/19/2020] [Indexed: 11/28/2022]
Abstract
Heart disease remains one of the significantcauses ofmortality and morbidity amongst the world's population. Predicting heart disease is considered as one of the vital issues in clinical data analysis. Since the number of data is rising gradually, it is muchcomplicatedforanalyzing and processing, and especially, it becomes difficult to maintain the e-healthcare data. Moreover, the prediction model under machine learning seems to be anessentialfacet in this research area. In this scenario, this paper aims to propose a new heart disease prediction model with the inclusion of specificprocesses like Feature Extraction, Record, Attribute minimization, and Classification. Initially, both statistical and higher-order statistical features are extracted under feature extraction. Subsequently, the record and attribute minimization carried out, where Component Analysis PCA plays its major role in solving the "curse of dimensionality."Finally, the prediction process takes place by the Neural Network (NN) model that intake the dimensionally reduced features. Moreover, the major intention of this paper deals with the accurate prediction. Hence, it is planned to influence the utility of meta-heuristic algorithms for the weight optimization of NN. This paper introduces a new hybrid algorithm termed Particle Swarm Optimization (PSO) merged LA update (PM-LU) algorithm that solves the above-mentioned optimization crisis, which hybrids the concept of Lion Algorithm (LA) and PSO algorithm. Finally, the efficiency of proposed work is compared over other conventional approaches and its superiority is proven with respect to certain performance measures. From the analysis, the presented PM-LU-NN scheme with regards to accuracy is 3.85%, 12.5%, 12.5%, 3.85%, and 7.41% better than LM-NN, WOA-NN, FF-NN, PSO-NN and LA-NN algorithms.
Collapse
Affiliation(s)
- Renji P Cherian
- Professor, Department of Computer Science & Engineering, Vimal Jyothi Engineering College, Chemperi, Kannur, India.
| | - Noby Thomas
- Assistant Professor, St. Joseph's College of Pharmacy, Cherthala, India.
| | - Sunder Venkitachalam
- Assistant Professor, Department of Computer Science & Engineering, Adi Shankara Institute of Engineering and Technology, Kalady, India.
| |
Collapse
|
11
|
Aggarwal Y, Das J, Mazumder PM, Kumar R, Sinha RK. Heart rate variability features from nonlinear cardiac dynamics in identification of diabetes using artificial neural network and support vector machine. Biocybern Biomed Eng 2020. [DOI: 10.1016/j.bbe.2020.05.001] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/03/2023]
|
12
|
Detection of sudden cardiac death by a comparative study of heart rate variability in normal and abnormal heart conditions. Biocybern Biomed Eng 2020. [DOI: 10.1016/j.bbe.2020.06.003] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/01/2023]
|
13
|
Swain SS, Patra D, Singh YO. Automated detection of myocardial infarction in ECG using modified Stockwell transform and phase distribution pattern from time-frequency analysis. Biocybern Biomed Eng 2020. [DOI: 10.1016/j.bbe.2020.06.004] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/24/2022]
|
14
|
Hussain L, Awan IA, Aziz W, Saeed S, Ali A, Zeeshan F, Kwak KS. Detecting Congestive Heart Failure by Extracting Multimodal Features and Employing Machine Learning Techniques. BIOMED RESEARCH INTERNATIONAL 2020; 2020:4281243. [PMID: 32149106 PMCID: PMC7049402 DOI: 10.1155/2020/4281243] [Citation(s) in RCA: 21] [Impact Index Per Article: 5.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 08/05/2019] [Revised: 12/21/2019] [Accepted: 01/20/2020] [Indexed: 01/11/2023]
Abstract
The adaptability of heart to external and internal stimuli is reflected by the heart rate variability (HRV). Reduced HRV can be a predictor of negative cardiovascular outcomes. Based on the nonlinear, nonstationary, and highly complex dynamics of the controlling mechanism of the cardiovascular system, linear HRV measures have limited capability to accurately analyze the underlying dynamics. In this study, we propose an automated system to analyze HRV signals by extracting multimodal features to capture temporal, spectral, and complex dynamics. Robust machine learning techniques, such as support vector machine (SVM) with its kernel (linear, Gaussian, radial base function, and polynomial), decision tree (DT), k-nearest neighbor (KNN), and ensemble classifiers, were employed to evaluate the detection performance. Performance was evaluated in terms of specificity, sensitivity, positive predictive value (PPV), negative predictive value (NPV), and area under the receiver operating characteristic curve (AUC). The highest performance was obtained using SVM linear kernel (TA = 93.1%, AUC = 0.97, 95% CI [lower bound = 0.04, upper bound = 0.89]), followed by ensemble subspace discriminant (TA = 91.4%, AUC = 0.96, 95% CI [lower bound 0.07, upper bound = 0.81]) and SVM medium Gaussian kernel (TA = 90.5%, AUC = 0.95, 95% CI [lower bound = 0.07, upper bound = 0.86]). The results reveal that the proposed approach can provide an effective and computationally efficient tool for automatic detection of congestive heart failure patients.
Collapse
Affiliation(s)
- Lal Hussain
- Department of Computer Science & IT, The University of Azad Jammu and Kashmir, City Campus, 13100 Muzaffarabad, Azad Kashmir, Pakistan
| | - Imtiaz Ahmed Awan
- Department of Computer Science & IT, The University of Azad Jammu and Kashmir, City Campus, 13100 Muzaffarabad, Azad Kashmir, Pakistan
| | - Wajid Aziz
- Department of Computer Science & IT, The University of Azad Jammu and Kashmir, City Campus, 13100 Muzaffarabad, Azad Kashmir, Pakistan
- College of Computer Sciences and Engineering, University of Jeddah, Jeddah 21959, Saudi Arabia
| | - Sharjil Saeed
- Department of Computer Science & IT, The University of Azad Jammu and Kashmir, City Campus, 13100 Muzaffarabad, Azad Kashmir, Pakistan
| | - Amjad Ali
- Department of Computer Science, COMSATS University Islamabad, Lahore Campus, Lahore, Pakistan
| | - Farukh Zeeshan
- Department of Computer Science, COMSATS University Islamabad, Lahore Campus, Lahore, Pakistan
| | - Kyung Sup Kwak
- Department of Information and Communication Engineering, Inha University, Incheon, Republic of Korea
| |
Collapse
|