An Automated High-Accuracy Detection Scheme for Myocardial Ischemia Based on Multi-Lead Long-Interval ECG and Choi-Williams Time-Frequency Analysis Incorporating a Multi-Class SVM Classifier

Cardiovascular Disease (CVD) is a primary cause of heart problems such as angina and myocardial ischemia. The detection of the stage of CVD is vital for the prevention of medical complications related to the heart, as they can lead to heart muscle death (known as myocardial infarction). The electrocardiogram (ECG) reflects these cardiac condition changes as electrical signals. However, an accurate interpretation of these waveforms still calls for the expertise of an experienced cardiologist. Several algorithms have been developed to overcome issues in this area. In this study, a new scheme for myocardial ischemia detection with multi-lead long-interval ECG is proposed. This scheme involves an observation of the changes in ischemic-related ECG components (ST segment and PR segment) by way of the Choi-Williams time-frequency distribution to extract ST and PR features. These extracted features are mapped to a multi-class SVM classifier for training in the detection of unknown conditions to determine if they are normal or ischemic. The use of multi-lead ECG for classification and 1 min intervals instead of beats or frames contributes to improved detection performance. The classification process uses the data of 92 normal and 266 patients from four different databases. The proposed scheme delivered an overall result with 99.09% accuracy, 99.49% sensitivity, and 98.44% specificity. The high degree of classification accuracy for the different and unknown data sources used in this study reflects the flexibility, validity, and reliability of this proposed scheme. Additionally, this scheme can assist cardiologists in detecting signal abnormality with robustness and precision, and can even be used for home screening systems to provide rapid evaluation in emergency cases.

[1]  H. S. Niranjana Murthy,et al.  Efficient Algorithm for Early Detection of Myocardial Ischemia using PCA based Features , 2016 .

[2]  Shaiful Jahari Hashim,et al.  Automated and High Accuracy Out-Of-Hospital Heart Diseases Early Detection System , 2017 .

[3]  Kosmas Dimitropoulos,et al.  Multi-lead ECG signal analysis for myocardial infarction detection and localization through the mapping of Grassmannian and Euclidean features into a common Hilbert space , 2019, Biomed. Signal Process. Control..

[4]  Gholamreza Attarodi,et al.  A New Pattern Recognition Method for Detection and Localization of Myocardial Infarction Using T-Wave Integral and Total Integral as Extracted Features from One Cycle of ECG Signal , 2014 .

[5]  Min-Ling Zhang,et al.  A Review on Multi-Label Learning Algorithms , 2014, IEEE Transactions on Knowledge and Data Engineering.

[6]  R. Rangayyan,et al.  Biomedical Signal Analysis , 2015 .

[7]  U. Rajendra Acharya,et al.  Automated detection of heart valve diseases using chirplet transform and multiclass composite classifier with PCG signals , 2020, Comput. Biol. Medicine.

[8]  Fan Li,et al.  A novel electrocardiogram parameterization algorithm and its application in myocardial infarction detection , 2015, Comput. Biol. Medicine.

[9]  Bo Norrving,et al.  Organizational update: the world health organization global status report on noncommunicable diseases 2014; one more landmark step in the combat against stroke and vascular disease. , 2015, Stroke.

[10]  Olaf Dössel,et al.  Comparison of Baseline Wander Removal Techniques considering the Preservation of ST Changes in the Ischemic ECG: A Simulation Study , 2017, Comput. Math. Methods Medicine.

[11]  Bor-Shyh Lin,et al.  Design of a Wearable 12-Lead Noncontact Electrocardiogram Monitoring System , 2019, Sensors.

[12]  Anthony Kaveh,et al.  Automated classification of coronary atherosclerosis using single lead ECG , 2013, 2013 IEEE Conference on Wireless Sensor (ICWISE).

[13]  Samarendra Dandapat,et al.  Multiscale Energy and Eigenspace Approach to Detection and Localization of Myocardial Infarction , 2015, IEEE Transactions on Biomedical Engineering.

[14]  Vahab Vahdat,et al.  Automating detection and localization of myocardial infarction using shallow and end-to-end deep neural networks , 2020, Appl. Soft Comput..

[15]  Yon-Kyu Park,et al.  A study on development of multi-parametric measure of heart rate variability diagnosing cardiovascular disease , 2007 .

[16]  A. Wingfield.  Treatment of Angina , 1954 .

[17]  U. Rajendra Acharya,et al.  Application of stacked convolutional and long short-term memory network for accurate identification of CAD ECG signals , 2018, Comput. Biol. Medicine.

[18]  Amit Kumar,et al.  Ischemia detection using Isoelectric Energy Function , 2016, Comput. Biol. Medicine.

[19]  Joerg F. Hipp,et al.  Time-Frequency Analysis , 2014, Encyclopedia of Computational Neuroscience.

[20]  Li Sun,et al.  ECG Analysis Using Multiple Instance Learning for Myocardial Infarction Detection , 2012, IEEE Transactions on Biomedical Engineering.

[21]  Witold Pedrycz,et al.  Ischemia episode detection in ECG using kernel density estimation, support vector machine and feature selection , 2012, Biomedical engineering online.

[22]  Ming Zhang,et al.  LPI Radar Waveform Recognition Based on Time-Frequency Distribution , 2016, Sensors.

[23]  U. Rajendra Acharya,et al.  Automated characterization and classification of coronary artery disease and myocardial infarction by decomposition of ECG signals: A comparative study , 2017, Inf. Sci..

[24]  Reza Tafreshi,et al.  Towards real-time detection of myocardial infarction by digital analysis of electrocardiograms , 2011, 2011 1st Middle East Conference on Biomedical Engineering.

[25]  Taskin Kavzoglu,et al.  A kernel functions analysis for support vector machines for land cover classification , 2009, Int. J. Appl. Earth Obs. Geoinformation.

[26]  Yao Chen,et al.  Application of Deep Learning Algorithm in Clinical Analysis of Patients With Serum Electrolyte Disturbance , 2020, IEEE Access.

[27]  Shaiful Jahari Hashim,et al.  Non-fiducial based ECG biometric authentication using one-class Support Vector Machine , 2017, 2017 Signal Processing: Algorithms, Architectures, Arrangements, and Applications (SPA).

[28]  H. Xian,et al.  Holiday and weekend effects on mortality for acute myocardial infarction in Shanxi, China: a cross-sectional study , 2020, International Journal of Public Health.

[29]  U. Rajendra Acharya,et al.  Automated diagnosis of congestive heart failure using dual tree complex wavelet transform and statistical features extracted from 2 s of ECG signals , 2017, Comput. Biol. Medicine.

[30]  Anastasios Tefas,et al.  Using robust dispersion estimation in support vector machines , 2013, Pattern Recognit..

[31]  Pawel Plawiak,et al.  Novel methodology of cardiac health recognition based on ECG signals and evolutionary-neural system , 2018, Expert Syst. Appl..

[32]  U. Rajendra Acharya,et al.  Analysis of Myocardial Infarction Using Discrete Wavelet Transform , 2010, Journal of Medical Systems.

[33]  Kaiyun Chen,et al.  Time–frequency analysis of nonstationary complex magneto-hydro-dynamics in fusion plasma signals using the Choi–Williams distribution , 2013 .

[34]  Lorenzo Sabattini,et al.  A Natural Infrastructure-Less Human–Robot Interaction System , 2017, IEEE Robotics and Automation Letters.

[35]  U. Rajendra Acharya,et al.  Automated detection of arrhythmias using different intervals of tachycardia ECG segments with convolutional neural network , 2017, Inf. Sci..

[36]  Jurandir Nadal,et al.  Reducing cross terms effects in the Choi-Williams transform of mioelectric signals , 2013, Comput. Methods Programs Biomed..

[37]  Md Billal Hossain,et al.  Using the Redundant Convolutional Encoder–Decoder to Denoise QRS Complexes in ECG Signals Recorded with an Armband Wearable Device , 2020, Sensors.

[38]  Xiaoyong Liu,et al.  Parameter optimization of support vector regression based on sine cosine algorithm , 2018, Expert Syst. Appl..

[39]  Ming Chen,et al.  CompareSVM: supervised, Support Vector Machine (SVM) inference of gene regularity networks , 2014, BMC Bioinformatics.

[40]  Zhitao Lin,et al.  Automatic QRS complex detection using two-level convolutional neural network , 2018, BioMedical Engineering OnLine.

[41]  William J. Williams,et al.  Improved time-frequency representation of multicomponent signals using exponential kernels , 1989, IEEE Trans. Acoust. Speech Signal Process..

[42]  Jason Weston,et al.  Multi-Class Support Vector Machines , 1998 .

[43]  Yujuan Si,et al.  A Robust Multilevel DWT Densely Network for Cardiovascular Disease Classification , 2020, Sensors.

[44]  Jing Zhang,et al.  Deep time-frequency representation and progressive decision fusion for ECG classification , 2020, Knowl. Based Syst..

[45]  W. J. Tompkins,et al.  Features Extraction of ECG signal for Detection of Cardiac Arrhythmias , 2020 .

[46]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[47]  Lu Cao,et al.  A New ECG Signal Classification Based on WPD and ApEn Feature Extraction , 2016, Circuits Syst. Signal Process..

[48]  Wan Azizun Wan Adnan,et al.  Performance Evaluation of Time-Frequency Distributions for ECG Signal Analysis , 2017, Journal of Medical Systems.

[49]  Boualem Boashash,et al.  Time-Frequency Signal Analysis and Processing: A Comprehensive Reference , 2015 .

[50]  Muhammad Arif,et al.  Detection and Localization of Myocardial Infarction using K-nearest Neighbor Classifier , 2012, Journal of Medical Systems.