Cardiac Arrhythmia Classification by Multi-Layer Perceptron and Convolution Neural Networks
<p>Ideal electrocardiogram (ECG) signal with key features indicated; (<b>a</b>) P wave, QRS complex, and T wave which play important roles in diagnosis abnormality of heart signal; (<b>b</b>) Features of an ECG signal; how and which part of heart is used to generates each feature [<a href="#B16-bioengineering-05-00035" class="html-bibr">16</a>].</p> "> Figure 2
<p>Comparison between simple neural network (NN) and deep NN; simple neural networks contain only one hidden layer as well as the input and output layers, while deep learning neural networks contain more than one hidden layer. In this case, there are four hidden layers between the input and output layers [<a href="#B17-bioengineering-05-00035" class="html-bibr">17</a>].</p> "> Figure 3
<p>System process flowchart of Multilayer Perceptron (MLP) and Convolution NN. To define features and labels in the dataset, two TensorFlow variables were defined. One hot encoder was used to encode the dataset.</p> "> Figure 4
<p>Proposed Algorithm of Convolution NN. Four convolutional layers were used. In addition, there is one output layer with a softmax layer to hold the output of the NN.</p> "> Figure 5
<p>The distribution of ECG segments used for training and testing. Eighty percent of the data was used for training and 20% was used for testing. Thirty percent of the training dataset was used for validation of the network.</p> "> Figure 6
<p>Accuracy and mean square error (MSE) graph of the MLP algorithm; (<b>a</b>) the accuracy of MLP increases as the number of epochs increases; (<b>b</b>) MSE reduces with every epoch and reaches the minimum point after 1000 epochs.</p> "> Figure 7
<p>Confusion matrix (CM) with and without normalization of the MLP algorithm; (<b>a</b>) 63 arrhythmias and 81 normal signals are correctly classified, while 9 arrhythmias and 12 normal ECG signals are misclassified; (<b>b</b>) CM with normalization gives an accuracy in percentage; in this case, accuracy was 88% for arrythmia and 87% for normal sinus ECG.</p> "> Figure 8
<p>Accuracy and MSE of the CNN algorithm; (<b>a</b>) accuracy of the CNN rises continuously with every epoch; (<b>b</b>) MSE of CNN reduces with each epoch.</p> "> Figure 9
<p>CM with and without normalization of the CNN algorithm; (<b>a</b>) most of the arrhythmia was correctly classified by CNN except bigeminy and FAV; those diseases might have the same characteristics as others; (<b>b</b>) normalized CM gives the accuracy of the CNN in percentage form.</p> ">
Abstract
:1. Introduction
2. Methodology
2.1. Problem Formulation
2.2. Convolutional Neural Network
2.3. Multilayer Perceptron
2.4. Model Architecture
2.5. ECG Data
2.6. Training of Data
2.7. Testing of Data
3. Results
4. Conclusions
Author Contributions
Conflicts of Interest
Appendix A
Class | Description | Example |
---|---|---|
Normal | Normal Sinus Rhythm means normal heart rate, in respect to both heart rate and rhythm. Heart Rate—60 to 100 BPM | |
VT | Ventricular Tachycardia is heart rhythm illness instigated by abnormal signals in the lower chambers of the heart. Heart Rate—More than 100 BPM | |
AFIB | Atrial Fibrillation is an irregular and fast heart rate than can increase chance of stroke, heart failure. Heart Rate—100 to 175 BPM | |
AF | Atrial Flutter is the same as AFIB. But, whereas AFIB causes increased heart rate without a regular pattern, AFL causes increased heart rate in a regular pattern. Heart Rate—100 to 175 BPM | |
SAV | Second Degree AV, is a disease of the cardiac conduction system in which the conduction of atrial impulse over the AV node and/or his bundle is delayed or blocked. | |
Bigeminy | Ventricular Bigeminy is a heart rhythm problem in which there is a continuous alternation of long and short heart beats. | |
FAV | In First Degree AV Block conduction is slowed, there are no missed beats. In first-degree AV block, every atrial impulse is transmitted to the ventricles, resulting in a regular ventricular rate. |
References
- Artis, S.G.; Mark, R.G.; Moody, G.B. Detection of atrial fibrillation using artificial neural networks. In Proceedings of the Computers in Cardiology, Venice, Italy, 23–26 September 1991; IEEE: Piscataway, NJ, USA, 1991; pp. 173–176. [Google Scholar]
- Melo, S.L.; Caloba, L.P.; Nadal, J. Arrhythmia analysis using artificial neural network and decimated electrocardiographic data. In Proceedings of the IEEE Conference on Computers in Cardiology, Cambridge, MA, USA, 24–27 September 2000; IEEE: Piscataway, NJ, USA, 2000; pp. 73–76. [Google Scholar]
- Moody, G.B.; Mark, R.G. The impact of the MIT-BIH arrhythmia database. IEEE Eng. Med. Biol. Mag. 2001, 20, 45–50. [Google Scholar] [CrossRef] [PubMed]
- Salam, A.K.; Srilakshmi, G. An algorithm for ECG analysis of arrhythmia detection. In Proceedings of the IEEE International Conference on Electrical, Computer and Communication Technologies (ICECCT), Coimbatore, India, 5–7 March 2015; IEEE: Piscataway, NJ, USA, 2015; pp. 1–6. [Google Scholar]
- Debbal, S.M. Model of differentiation between normal and abnormal heart sounds in using the discrete wavelet transform. J. Med. Bioeng. 2014, 3, 5–11. [Google Scholar] [CrossRef]
- Perez, R.R.; Marques, A.; Mohammadi, F. The application of supervised learning through feed-forward neural networks for ECG signal classification. In Proceedings of the IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), Vancouver, BC, Canada, 15–18 May 2016; IEEE: Piscataway, NJ, USA, 2016; pp. 1–4. [Google Scholar]
- Palreddy, S.; Tompkins, W.J.; Hu, Y.H. Customization of ECG beat classifiers developed using SOM and LVQ. In Proceedings of the IEEE 17th Annual Conference on Engineering in Medicine and Biology Society, Montreal, QC, Canada, 20–23 September 1995; IEEE: Piscataway, NJ, USA, 1995; pp. 813–814. [Google Scholar]
- Elsayad, A.M. Classification of ECG arrhythmia using learning vector quantization neural networks. In Proceedings of the 2009 International Conference on Computer Engineering & Systems, Cairo, Egypt, 14–16 December 2009; IEEE: Piscataway, NJ, USA, 2009; pp. 139–144. [Google Scholar]
- Gautam, M.K.; Giri, V.K. A neural network approach and wavelet analysis for ECG classification. In Proceedings of the 2016 IEEE International Conference on Engineering and Technology (ICETECH), Coimbatore, India, 17–18 March 2016; IEEE: Piscataway, NJ, USA, 2016; pp. 1136–1141. [Google Scholar]
- Zebardast, B.; Ghaffari, A.; Masdari, M. A new generalized regression artificial neural networks approach for diagnosing heart disease. Int. J. Innov. Appl. Stud. 2013, 4, 679. [Google Scholar]
- Acharya, U.R.; Fujita, H.; Lih, O.S.; Adam, M.; Tan, J.H.; Chua, C.K. Automated detection of coronary artery disease using different durations of ECG segments with convolutional neural network. Knowl. Based Syst. 2017, 132, 62–71. [Google Scholar] [CrossRef]
- Nilanon, T.; Yao, J.; Hao, J.; Purushotam, S.; Liu, Y. Normal/abnormal heart recordings classification by using convolutional neural network. In Proceedings of the IEEE Conference on Computing in Cardiology Conference (CinC), Vancouver, BC, Canada, 11–14 September 2016; IEEE: Piscataway, NJ, USA, 2016; pp. 585–588. [Google Scholar]
- Jigar, A.D. Swift single image super resolution using deep convolution neural network. In Proceedings of the International Conference on Communication and Electronics Systems (ICCES), Coimbatore, India, 21–22 October 2016; IEEE: Piscataway, NJ, USA, 2016; pp. 1–6. [Google Scholar]
- Fernando, A.; Oliver, C.; Marco, A.F.P.; Adam, M.; de Maarten, V. Comparing feature based classifiers and convolutional neural networks to detect arrhythmia from short segments of ECG. In Proceedings of the Conference on Computing in Cardiology (CinC), Rennes, France, 24–27 September 2017; IEEE: Piscataway, NJ, USA, 2017. [Google Scholar]
- Shadi, G.; Mostafa, A.; Nasimalsadat, M.; Kamran, K.; Ali, G. Atrial fibrillation detection using feature based algorithm and deep conventional neural network. In Proceedings of the Conference on Computing in Cardiology (CinC), Rennes, France, 24–27 September 2017; IEEE: Piscataway, NJ, USA, 2017. [Google Scholar]
- Chow, G.V.; Marine, J.E.; Fleg, J.L. Epidemiology of arrhythmias and conduction disorders in older adults. Clin. Geriatr. Med. 2012, 28, 539–553. [Google Scholar] [CrossRef] [PubMed]
- Acharya, U.R.; Fujita, H.; Adam, M.; Oh, S.L.; Tan, J.H.; Sudarshan, V.K.; Koh, J.E.W. Automated characterization of arrhythmias using nonlinear features from tachycardia ECG beats. In Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics (SMC), Budapest, Hungary, 9–12 October 2016; IEEE: Piscataway, NJ, USA, 2016. [Google Scholar]
- Desai, U.; Martis, R.J.; Acharya, U.R.; Nayak, C.G.; Seshikala, G.; Shettyk, R. Diagnosis of multiclass tachycardia beats using recurrence quantification analysis and ensemble classifiers. J. Mech. Med. Biol. 2016, 16. [Google Scholar] [CrossRef]
- Zubair, M.; Kim, J.; Yoon, C.W. An automated ECG beat classification system using convolutional neural networks. In Proceedings of the 6th International Conference on IT Convergence and Security (ICITCS), Prague, Czech Republic, 26–29 September 2016; IEEE: Piscataway, NJ, USA, 2016; pp. 1–5. [Google Scholar]
- Oquab, M.; Bottou, L.; Laptev, I.; Sivic, J. Is object localization for free?—Weakly-supervised learning with convolutional neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA, 12 June 2015; IEEE: Piscataway, NJ, USA, 2016; pp. 685–694. [Google Scholar]
- Krizhevsky, A.; Sutskever, I.; Hinton, G.E. ImageNet classification with deep convolutional neural networks. In Proceedings of the Neural Information Processing Systems 2012, Lake Tahoe, NV, USA, 3–8 December 2012; The MIT Press: Cambridge, MA, USA, 2012; pp. 1097–1105. [Google Scholar]
- Kiranyaz, S.; Ince, T.; Gabbouj, M. Real-time patient-specific ECG classification by 1-D convolutional neural network. IEEE Trans. Biomed. Eng. 2016, 63, 664–675. [Google Scholar] [CrossRef] [PubMed]
- Fahim, S.; Khalil, I. Diagnosis of cardiovascular abnormalities from compressed ECG: A datamining-based approach. IEEE Trans. Inf. Technol. Biomed. 2011, 15, 33–39. [Google Scholar]
- Ciresan, D.C.; Meier, U.; Gambardella, L.M.; Schumacher, J. Convolutional neural network committees for handwritten character classification. In Proceedings of the 2011 International Conference on Document Analysis and Recognition (ICDAR), Beijing, China, 18–21 September 2011; IEEE: Piscataway, NJ, USA, 2011; pp. 1135–1139. [Google Scholar]
- Sainath, T.N.; Mohamed, A.-R.; Kingsbury, B.; Ramabhadran, B. Deep convolutional neural networks for lvcsr. In Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Vancouver, BC, Canada, 26–31 May 2013; IEEE: Piscataway, NJ, USA, 2011; pp. 8614–8618. [Google Scholar]
- Golkov, V.; Dosovitskiy, A.; Sperl, J.I.; Menzel, M.I.; Czisch, M.; Sämann, P.; Brox, T.; Cremers, D. q-Space deep learning: Twelve-fold shorter and model-free diffusion MRI scans. IEEE Trans. Med. Imaging 2016, 35, 1344–1351. [Google Scholar] [CrossRef] [PubMed]
- Martis, R.J.; Acharya, U.R.; Adeli, H.; Prasad, H.; Tan, J.H.; Chua, K.C.; Too, C.L.; Yeo, S.W.J.; Tong, L. Computer-aided diagnosis of atrial arrhythmia using dimensionality reduction methods on transform domain representation. Biomed. Signal Process. Control 2014, 13, 295–305. [Google Scholar] [CrossRef]
- Martis, R.J.; Acharya, U.R.; Prasad, H.; Chua, K.C.; Lim, C.M.; Suri, J.S. Application of higher order statistics for atrial arrhythmia classification. Biomed. Signal Process. Control 2013, 8, 888–900. [Google Scholar] [CrossRef]
- Wang, Y.; Zhu, Y.S.; Thakor, N.V.; Xu, Y.H. A short-time multifractal approach for arrhythmia detection based on fuzzy neural network. IEEE Trans. Biomed. Eng. 2001, 48, 989–995. [Google Scholar] [CrossRef] [PubMed]
- Yan, Z.; Zhan, Y.; Peng, Z.; Liao, S.; Shinagawa, Y.; Zhang, S.; Metaxas, D.N.; Zhou, X.S. Multi-instance deep learning: Discover discriminative local anatomies for body part recognition. IEEE Trans. Med. Imaging 2016, 35, 1332–1343. [Google Scholar] [CrossRef] [PubMed]
- Najarian, K.; Splinter, R. Biomedical Signal and Image Processing, 2nd ed.; CRC Press, Taylor and Francis Group: Boca Raton, FL, USA, 2012. [Google Scholar]
- He, K.; Zhang, X.; Ren, S.; Sun, J. Identity mappings in deep residual networks. In Proceedings of the 14th European Conference on Computer Vision, Cham, The Netherlands, 8–16 October 2016; Springer: Berlin, Germany, 2016; pp. 630–645. [Google Scholar]
- Clifford, G.D.; Liu, C.Y.; Moody, B.; Lehman, L.; Silva, I.; Li, Q.; Johnson, A.E.W.; Mark, R.G. AF classification from a short single lead ECG recording: The PhysioNet computing in cardiology challenge 2017. In Proceedings of the Conference on Computing in Cardiology (CinC), Rennes, France, 24–27 September 2017; IEEE: Piscataway, NJ, USA, 2017. [Google Scholar]
- Burges, C.J.C. A tutorial on support vector machines for pattern recognition. Data Min Knowl Discov. 1998, 2, 121–167. [Google Scholar] [CrossRef]
- Vapnik, V. The Nature of Statistical Learning Theory; Springer: New York, NY, USA, 2013. [Google Scholar]
Layers | Type | Size of Neurons (Output Layer) | Filter Size of Each Layer |
---|---|---|---|
0–1 | Convolution | (None, 1, 60, 1) | 32 |
1–2 | Max Pooling | (None, 1, 30, 1) | 2 |
2–3 | Convolution | (None, 1, 30, 1) | 32 |
3–4 | Max Pooling | (None, 1, 15, 1) | 2 |
4–5 | Convolution | (None, 1, 15, 1) | 32 |
5–6 | Max Pooling | (None, 1, 8, 1) | 2 |
6–7 | Convolution | (None, 1, 8, 1) | 32 |
5–6 | Fully connected layer | 2048 | - |
© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Savalia, S.; Emamian, V. Cardiac Arrhythmia Classification by Multi-Layer Perceptron and Convolution Neural Networks. Bioengineering 2018, 5, 35. https://doi.org/10.3390/bioengineering5020035
Savalia S, Emamian V. Cardiac Arrhythmia Classification by Multi-Layer Perceptron and Convolution Neural Networks. Bioengineering. 2018; 5(2):35. https://doi.org/10.3390/bioengineering5020035
Chicago/Turabian StyleSavalia, Shalin, and Vahid Emamian. 2018. "Cardiac Arrhythmia Classification by Multi-Layer Perceptron and Convolution Neural Networks" Bioengineering 5, no. 2: 35. https://doi.org/10.3390/bioengineering5020035
APA StyleSavalia, S., & Emamian, V. (2018). Cardiac Arrhythmia Classification by Multi-Layer Perceptron and Convolution Neural Networks. Bioengineering, 5(2), 35. https://doi.org/10.3390/bioengineering5020035