- Research
- Open Access
Robust algorithm for arrhythmia classification in ECG using extreme learning machine
- Jinkwon Kim^{1},
- Hang Sik Shin^{1},
- Kwangsoo Shin^{1} and
- Myoungho Lee^{1}Email author
https://doi.org/10.1186/1475-925X-8-31
© Kim et al; licensee BioMed Central Ltd. 2009
- Received: 19 February 2009
- Accepted: 28 October 2009
- Published: 28 October 2009
Abstract
Background
Recently, extensive studies have been carried out on arrhythmia classification algorithms using artificial intelligence pattern recognition methods such as neural network. To improve practicality, many studies have focused on learning speed and the accuracy of neural networks. However, algorithms based on neural networks still have some problems concerning practical application, such as slow learning speeds and unstable performance caused by local minima.
Methods
In this paper we propose a novel arrhythmia classification algorithm which has a fast learning speed and high accuracy, and uses Morphology Filtering, Principal Component Analysis and Extreme Learning Machine (ELM). The proposed algorithm can classify six beat types: normal beat, left bundle branch block, right bundle branch block, premature ventricular contraction, atrial premature beat, and paced beat.
Results
The experimental results of the entire MIT-BIH arrhythmia database demonstrate that the performances of the proposed algorithm are 98.00% in terms of average sensitivity, 97.95% in terms of average specificity, and 98.72% in terms of average accuracy. These accuracy levels are higher than or comparable with those of existing methods. We make a comparative study of algorithm using an ELM, back propagation neural network (BPNN), radial basis function network (RBFN), or support vector machine (SVM). Concerning the aspect of learning time, the proposed algorithm using ELM is about 290, 70, and 3 times faster than an algorithm using a BPNN, RBFN, and SVM, respectively.
Conclusion
The proposed algorithm shows effective accuracy performance with a short learning time. In addition we ascertained the robustness of the proposed algorithm by evaluating the entire MIT-BIH arrhythmia database.
Keywords
- Support Vector Machine
- Hide Layer
- Extreme Learn Machine
- Hide Neuron
- Radial Basis Function Network
Background
Arrhythmia is a form of heart conduction system disease that causes an inefficient heart beat. Typically, arrhythmia is diagnosed through an electrocardiogram procedure. Because arrhythmia represents abrupt and abnormal ECG beats, physicians diagnose arrhythmia based on long-term ECG data using an ECG recording system like the Holter recorder. In addition, various remote and mobile healthcare systems that are adapting ECG recorders are increasing in number these days, and the importance of an automatic arrhythmia classification algorithm is being increasingly recognized. There are many existing studies on the classification of arrhythmia, and the algorithm is generally composed of the pre-processing part, the feature extraction part, and the classification part.
The pre-processing part removes noise components and does other forms of processing for more accurate feature extraction or classification. The main noise components of ECG include baseline drift, power line interference, and moving artefacts [1]. Various reports have been published on the filtering methods for removing noise components while preserving both ECG morphology and fast processing.
The feature extraction part makes feature vectors that are used later in the classification part. Various signal compression algorithms are used to represent the signal's characteristics efficiently with a small computational burden. Many transformation methods, such as principal component analysis (PCA), or independent component analysis (ICA)[2], are very frequently used. The classification part makes an arrhythmia diagnosis using acquired feature vectors from the feature extraction part. Statistical approaches [3], fuzzy inference approaches, and neural network approaches [4] are the typically used methods for ECG pattern classification. The statistical approaches classify ECG patterns by using statistical modelling, which is acquired from the data. However, in the case of statistical approaches, there are some difficulties in acquiring many types of data and in selecting a model that best represents data distribution. The fuzzy inference approaches have less computational burden than others, but those approaches have a subjective nature because membership function selection is accomplished by applying the opinions of experts and repeated experimentation. In addition, the complex decision region acquired through the machine learning approach is considered as one of the neural network approaches. However, a huge learning dataset, large computational burden and extended learning time are pointed out as the main shortcomings of the neural network approaches.
There have been many studies that found ways to overcome the weaknesses of the neural network approaches. Most of the recent research projects on improving the arrhythmia classification algorithm are classified into two types of approaches. One approach involves finding a better ECG feature extraction method, such as dimension reduction, and the other is concerned with finding a better classifier. Many studies on ECG feature extraction have been reported using PCA and ICA for dimension reduction; and Fourier transform [5] and Wavelet transform [4, 6] for frequency component representation. There are also many studies that adapted a new classifier such as a Support Vector Machine (SVM)[7]. Some comparative studies of various data reduction[8–10], feature extraction[10, 11], and classification methods[11] were presented recently, but the size of test data set was relatively small.
As a kind of classifier, Extreme Learning Machine (ELM) was able to overcome the difficulties of a neural network through a fast learning speed and high performance [12]; the problems that persisted were an extended learning time for a gradient-based learning algorithm, the possibility of local minima converging, and a degraded performance due to overtraining. A BPNN that is typically used as a learning algorithm on a Single Hidden Layer Feedforward Neural Network (SLFN) adjusts the weights between layers based on propagating errors from the output layer to the input layer. However, it was demonstrated that a SLFN classifier that has randomly selected weights between the input layer and the hidden layer may have the ability to classify the data set only with the controlling weights between the hidden layer and the output layer [13]. Based on this research result, the ELM acquired optimal weights between the hidden layer and the output layer analytically, with randomly selected weights between the input layer and the hidden layer.
The performance of the classification not only depends on the classifier, but also depends on the features, and better ECG signal processing is of great benefit to feature extraction. Thus in this study, we proposed and analyzed the proper signal processing methods for each part of arrhythmia classification algorithm: preprocessing, feature extraction, and classification part, in an effort to develop robust algorithm. We used Morphology Filter (MF) as the pre-processing part to remove the noise component while preserving ECG morphology, and time domain features and morphology features of ECG compressed by PCA as the feature extraction part. For the classification part we used ELM in order to reduce learning time while maintaining high accuracy. We first evaluated the each component of the proposed arrhythmia classification algorithm separately, and made the comparative study of the performances of the algorithm using an ELM, NN, RBFN, or SVM as the classifier. In addition, the performance of proposed algorithm was compared with that of other researches.
Methods
Morphology Filtering
where '⊕ ' and '⊖ ', dilation and erosion respectively, are morphological operators. B_{2} is the structural element [0,0,0,0,0] to remove the high frequency noise. And, B_{1}, which is a structural element [0,1,5,1,0], compensates for the high frequency components of the original signal, which are removed by B_{2}
Principal Component Analysis
The bases from Equation (4) are eigenvectors of the scatter matrix. To maximize distribution of projected data in d' dimension, we have to select d' bases from Equation (4) in the order of their eigenvalues. The scatter matrix S is real and symmetric, thus the eigenvectors e_{ i }are mutually orthogonal. The reduced dimensional space is made up of selected eigenvectors.
Extreme Learning Machine
By using the Moor-Penrose Generalized Inverse of H, the optimal β has the minimum norm least-squares solution of the linear system. With this process we can establish two effects. The first effect is that we can take a minimum error condition, because obtained is least-squares solution. In addition, optimal is not only the least-squares solution but also the minimum norm among these solution. Thus, ELM has a better generalization performance than a typical BPNN [15].
Back Propagation Neural Network
BPNN takes a long time to learn and has the risk of falling into a local minimum, because it solves the Equation (9) iteratively until J(w) reaches the minimum.
Radial Basis Function Network
where d_{ max }is the maximum distance between any 2 centers, and m is the number of centers. Finally, w are computed by means of the pseudo-inverse method.
Support Vector Machine
SVM is basically a binary classifier. But in this paper, we used a one-against-all multiclass SVM because of many classes for each arrhythmia.
Experimental Method
The proposed algorithm was evaluated using the MIT-BIH arrhythmia database [18]. The evaluation with the entire MIT-BIH arrhythmia database shows the robustness of proposed algorithm. (102, 104, and 114 ECG data files in the MIT-BIH arrhythmia database were excluded since those files are not recorded on Modified Lead II.) We selected major beat types with a coverage ratio exceeding 1% in the entire MIT-BIH arrhythmia database. By doing that, we had 6 major beat types. They are normal beats, left bundle branch block (LBBB) beats, right bundle branch block (RBBB) beats, premature ventricular contractions (PVC), atrial premature beats (APB), and paced beats (PB). The total of the coverage ratios of those 6 major beat types is 97.76% as is shown in Table 1.
Composition of the data set
Beat type | Number of beats in Test data set | Number of beats in Training data set | Possession rate (%) |
---|---|---|---|
Normal | 54516 | 5000 | 70.2573 |
LBBB | 6036 | 1000 | 7.7746 |
RBBB | 5410 | 1000 | 6.9889 |
PVC | 5293 | 1000 | 6.8166 |
APB | 1896 | 1000 | 2.4407 |
PB | 2702 | 1000 | 3.4853 |
Total | 75853 | 10000 | 97.76 |
The constructed ECG dataset was filtered by a MF. Then, the current RR interval (RRI), the ratio of the current RRI against the next RRI (RRIR), the ratio of the current RRI against the average of the latest 10 beats (10RRIR), and the R peak amplitude (Ramp) were extracted from the filtered signal as ECG descriptive features. The RRIR was introduced to reflect the compensatory pause of the APB and PVC, and the 10RRIR compensates for the variation in the normal RR interval range. Besides ECG descriptive features, morphology features were acquired from ECG data around an R peak (250 ms, 90 samples), which is provided by the databases' annotation. The morphology features compressed by PCA and the descriptive features were included in the feature vectors to train and test ELM. The evaluation was performed with Matlab R2007b and with Intel Quad core 2.40 GHz, 2 GB RAM, Microsoft windows XP platform.
Results
This section presents the results of the analysis of each component. Concerning the morphology filtering component, we compared MF with a typical filtering method in terms of preserving ECG morphology, and for the feature extraction component we analyzed the distribution of features in each beat type to estimate the efficiency of the features. Finally, for the classification component, we simulated the algorithm in terms of sensitivity, specificity, and the accuracy basis by varying the number of principal components and hidden neurons, in order to evaluate the optimal ELM structure. The obtained optimal results of the proposed algorithm were compared with those of other existing research papers in terms of the size of the training dataset, test dataset, and accuracy. The training time, which is another advantage of the proposed algorithm, was also evaluated with BPNN, RBF, or SVM based algorithm.
Result of Morphology Filtering
Distribution of Feature Set Results
Results of the Proposed Algorithm
where TP, TN, FP, and FN are True Positive, True Negative, False Positive, and False Negative, respectively.
Results of the proposed algorithm with 720 hidden neurons and 14 components
Results from Proposed Algorithm | Actual heart beat type | ||||||
---|---|---|---|---|---|---|---|
Nor | LBBB | RBBB | PVC | APB | PB | Total | |
Normal | 53885 | 90 | 35 | 276 | 188 | 1 | 54475 |
LBBB | 83 | 5922 | 2 | 64 | 5 | 0 | 6076 |
RBBB | 65 | 1 | 5356 | 29 | 5 | 0 | 5456 |
PVC | 104 | 16 | 1 | 4783 | 6 | 0 | 4910 |
APB | 379 | 7 | 16 | 141 | 1692 | 1 | 2236 |
PB | 0 | 0 | 0 | 0 | 0 | 2700 | 2700 |
Total | 54516 | 6036 | 5410 | 5293 | 1896 | 2702 | 75853 |
Accuracy, Sensitivity, Specificity (%) | |||||||
Beat types | Nor | LBBB | RBBB | PVC | APB | PB | Avg. |
sensitivity | 98.84 | 98.11 | 99.00 | 90.36 | 89.24 | 99.93 | 98.00 |
specificity | 97.23 | 99.78 | 99.86 | 99.82 | 99.26 | 100.00 | 97.95 |
accuracy | 98.39 | 99.65 | 99.80 | 99.16 | 99.01 | 100.00 | 98.72 |
Evaluation of the learning times and the testing accuracies among algorithms using ELM with 10 principle components
The number of hidden neurons | 200 | 400 | 600 | 800 | 1000 | 1200 | 1400 | 1600 | 1800 | 2000 |
---|---|---|---|---|---|---|---|---|---|---|
Training time (seconds) | 0.95 | 3.13 | 7.09 | 13.61 | 22.48 | 35.13 | 50.47 | 70.11 | 94.30 | 123.77 |
Testing accuracy (%) | 97.76 | 98.50 | 98.60 | 98.72 | 98.72 | 98.60 | 98.37 | 98.19 | 98.06 | 97.75 |
Evaluation of the learning times and the testing accuracies among algorithms using BPNN with 10 principle components
Learning rate η | The number of hidden neurons | 10 | 20 | 30 | 40 | 50 | 60 | 70 | 80 | 90 | 100 |
---|---|---|---|---|---|---|---|---|---|---|---|
0.0005 | Training time (seconds) | 4092.19 | 4213.25 | 4276.55 | 4424.77 | 4455.69 | 4589.70 | 4553.02 | 4640.63 | 4745.16 | 4884.30 |
Testing accuracy (%) | 94.77 | 94.00 | 94.16 | 95.32 | 94.74 | 92.62 | 92.52 | 90.28 | 88.77 | 89.91 | |
0.001 | Training time (seconds) | 4058.33 | 4157.41 | 4218.09 | 4354.41 | 4408.23 | 4500.06 | 5.75* | 1532.38* | 2.38* | 32.61* |
Testing accuracy (%) | 95.02 | 94.45 | 95.83 | 95.51 | 93.56 | 91.30 | 78.17 | 78.17 | 78.17 | 78.17 | |
0.002 | Training time (seconds) | 4054.50 | 4212.85 | 4213.71 | 4385.40 | 1.11* | 1.12* | 1.12* | 3.44* | 1.17* | 144.12* |
Testing accuracy (%) | 96.42 | 95.72 | 89.37 | 86.15 | 78.17 | 78.17 | 78.17 | 78.17 | 78.17 | 78.17 |
Evaluation of the learning times and the testing accuracies among algorithms using RBFN with 10 principle components
The number of hidden neurons | 200 | 400 | 600 | 800 | 1000 | 1200 | 1400 | 1600 | 1800 | 2000 |
---|---|---|---|---|---|---|---|---|---|---|
Training time (seconds) | 46.03 | 220.78 | 290.30 | 392.64 | 461.19 | 852.83 | 972.56 | 1426.38 | 1208.97 | 2008.20 |
Testing accuracy (%) | 86.00 | 87.81 | 88.61 | 88.24 | 87.40 | 89.16 | 89.36 | 89.31 | 89.21 | 89.37 |
Evaluation of the learning times and the testing accuracies among algorithms using SVM with 10 principle components
C | 1 | 1000 | 2000 | 3000 | 4000 | 5000 | 6000 | 7000 | 8000 | 9000 |
---|---|---|---|---|---|---|---|---|---|---|
Training time (seconds) | 98.95 | 52.73 | 46.83 | 47.67 | 39.73 | 50.84 | 40.25 | 37.78 | 45.89 | 39.03 |
Testing accuracy (%) | 88.89 | 97.71 | 98.03 | 98.19 | 98.28 | 98.35 | 98.39 | 98.40 | 98.46 | 98.48 |
In the algorithm based on RBF, we evaluated by varying the number of hidden neurons from 200 to 2000. The accuracy converged to 89% and showed no more improvement after the number of hidden neurons reached 1200. The spread of RBF was chosen by the normalization as Equation (11). The performance of SVM is based on the type of kernel and C which controls a weight between the size of margin and the number of misclassified data. We selected the Gaussian kernel like RBF and the sigma of Gaussian kernel is fixed as 2. The C was varied between 1 and 9000, and the accuracy converged to 98% when C is over 2000.
Discussion
We evaluated the proposed algorithm in the pre-processing (filtering), feature extraction and classification parts. The proposed algorithm adapted MF as the preprocessing part. 4 ECG descriptive features and morphology features compressed by PCA made up the feature vectors. Since MF is superior in performance to frequency based filtering in terms of preserving ECG morphology, it is thought that MF will become an attractive method concerning the ECG arrhythmia classification algorithm. The distribution characteristics of the proposed ECG descriptive features, especially RRIR and some PCs, showed that they contained effective information.
The proposed algorithm represented the advantages of a fast learning speed and high accuracy in comparison with other gradient based learning algorithms, and was evaluated as having a large dataset. The ELM based algorithm showed the most accurate performance, the SVM based one was slightly lower, but the BPNN and RBF based algorithms were fairly low in terms of accuracy based on when the proposed algorithm and other classifiers based algorithms reached the maximum accuracy with 10 principal components. But the learning time of the proposed algorithm was shorter about 290, 70, and 3 times than that of BPNN, RBF, and SVM based algorithm.
These results demonstrate that the proposed algorithm used more hidden neurons than did the BPNN based algorithm. It reflected the characteristic of ELM that used a portion of the hidden neurons among the randomly setting weights between the input layer and the hidden layer. In addition, we found a tendency for performance to decrease when using over 1800 hidden neurons. This phenomenon was caused by the minimum norm least-squares solution of the linear system in Equation (7) through the Moor-Penrose Generalized Inverse of H. The minimum norm least-squares solution searches for the least-squares solution first, and then searches for the minimum norm among those solutions Therefore, the accuracy of the algorithm concerning the training dataset increases as the number of hidden neurons increases, but the generalization performance decreases when too many hidden neurons are used. Nevertheless, ELM is much easier to decide and evaluate the structure of neural network than BPNN.
Comparing the results of preceding research
Method | Number of beat types Training Data Set/Test Data Set | Accuracy (%) |
---|---|---|
ICA-BPNN[2] | 8 (4900/4900) | 98.37 |
DWT-NN[4] | 13 (30293/75130) | 96.79 |
FTNN [5] | 3 (540/250) | 98.0 |
Wavelet-PNN[6] | 6 (11600/11600) | 99.65 |
MOE [21] | 4 about (10000/49260) | 94.0 |
Fhyb-HOSA[22] | 7 (4035/3150) | 96.06 |
BSS-Fourier[23] | 5 (320/160)* | 85.04 |
PCA-ELM[24] | 7 (3450/3450) | 97.45 |
Proposed Algorithm | 6 (10000/75853) | 98.72 |
Conclusion
We proposed an arrhythmia classification algorithm using ELM in ECG. The proposed algorithm showed effective accuracy performance with a short learning time. In addition, we ascertained the robustness of the proposed algorithm by evaluating the entire MIT-BIH arrhythmia database. All beat types were classified with high accuracy, but in APB the sensitivity was slightly lower than 90%. This result was due to the characteristic of one patient in the MIT-BIH arrhythmia database who had bradycardia and APB at the same time. Thus, the features of that patient's APB beat are similar to the features of other patients' normal beats. Our future direction is to develop a feature set to manage various situations like the one mentioned above, and to separate the subjects included in the training dataset or in the test dataset for better generalization performance.
Declarations
Authors’ Affiliations
References
- Friesen GM, Jannett TC, Jadallah MA, Yates SL, Quint SR: A Comparison of the Noise Sensitivity of Nine QRS Detection Algorithms. IEEE Trans Biomed Eng 1990, 37: 85–98.View ArticleGoogle Scholar
- Yu SN, Chou KT: Combining independent component analysis and backpropagation neural network for ECG beat classification. Conf Proc IEEE Eng Med Biol Soc 2006, 1: 3090–3093.View ArticleGoogle Scholar
- Ge D, Srinivasan N, Krishnan SM: Cardiac arrhythmia classification using autoregressive modeling. Biomed Eng Online 2002, 1: 5.View ArticleGoogle Scholar
- Prasad GK, Sahambi JS: Classification of ECG arrhythmias using multi-resolution analysis and neural networks. Conf Proc IEEE Convergent Tech 2003, 1: 227–231.Google Scholar
- Minami K, Nakajima H, Toyoshima T: Real-time discrimination of ventricular tachyarrhythmia with Fourier-transform neural network. IEEE Trans Biomed Eng 1999, 46: 179–185.View ArticleGoogle Scholar
- Yu SN, Chen YH: Electrocardiogram beat classification based on wavelet transformation and probabilistic neural network. Pattern Recognition Letters 2007, 28: 1142–1150.View ArticleGoogle Scholar
- Song MH, Lee J, Cho SP, Lee KJ, Yoo SK: Support Vector Machine Based Arrhythmia Classification Using Reduced Features. International Journal of Control, Automation, and Systems 2005, 3: 571–579.Google Scholar
- Ozbay Y, Ceylan R, Karlik B: A fuzzy clustering neural network architecture for classification of ECG arrhythmias. Comput Biol Med 2006, 36: 376–388.View ArticleGoogle Scholar
- Ozbay Y, Ceylan R, Karlik B: A Novel Approach for Classification of ECG Arrhythmias: Type-2 Fuzzy Clustering Neural Network. Expert Syst Appl 2009, 36: 6721–6726.View ArticleGoogle Scholar
- Ceylan R, Ozbay Y: Comparison of FCM, PCA and WT techniques for classification ECG arrhythmias using artificial neural network. Expert Syst Appl 2007, 33: 286–295.View ArticleGoogle Scholar
- Ubeyli ED: Usage of eigenvector methods in implementation of automated diagnostic systems for ECG beats. Digit Signal Process 2008, 18: 33–48.View ArticleGoogle Scholar
- Huang GB, Zhu QY, Siew CK: Extreme learning machine: Theory and applications. Neurocomputing 2006, 70: 489–501.View ArticleGoogle Scholar
- Huang GB, Zhu QY, Siew CK: Extreme Learning Machine: A New Learning Scheme of Feedforward Neural Networks. Conf Proc IEEE Neural Networks 2004, 2: 985–990.Google Scholar
- Sun Y, Chan K, Krishnan SM: ECG signal conditioning by morphological filtering. Comput Biol Med 2002, 32: 465–479.View ArticleGoogle Scholar
- Bartlett PL: The sample complexity of pattern classification with neural networks: The size of the weights is more important than the size of the network. IEEE Trans Information Theory 1998, 44: 525–536.MathSciNetView ArticleGoogle Scholar
- Duda RO, Hart PE, Stork DG: Pattern classification. 2nd edition. New York: Wiley; 2001.Google Scholar
- Cristianini N, Shawe-Taylor J: An introduction to support vector machines: and other kernel-based learning methods. Cambridge; New York: Cambridge University Press; 2000.View ArticleGoogle Scholar
- MIT-BIH arrhythmia database[http://www.physionet.org/physiobank/database/mitdb/]
- Efron B, Tibshirani R: An introduction to the bootstrap. New York: Chapman & Hall; 1993.View ArticleGoogle Scholar
- Christov I, Bortolan G: Ranking of pattern recognition parameters for premature ventricular contractions classification by neural networks. Physiol Meas 2004, 25: 1281–1290.View ArticleGoogle Scholar
- Hu YH, Palreddy S, Tompkins WJ: A patient-adaptable ECG beat classifier using a mixture of experts approach. IEEE Trans Biomed Eng 1997, 44: 891–900.View ArticleGoogle Scholar
- Osowski S, Linh TH: ECG beat recognition using fuzzy hybrid neural network. IEEE Trans Biomed Eng 2001, 48: 1265–1271.View ArticleGoogle Scholar
- Owis MI, Youssef AB, Kadah YM: Characterisation of electrocardiogram signals based on blind source separation. Med Biol Eng Comput 2002, 40: 557–564.View ArticleGoogle Scholar
- Kim J, Shin H, Lee Y, Lee M: Algorithm for classifying arrhythmia using Extreme Learning Machine and principal component analysis. Conf Proc IEEE Eng Med Biol Soc 2007, 2007: 3257–3260.Google Scholar
Copyright
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.