Gender Classification from ECG Signal Analysis using Least Square Support Vector Machine

In this present paper it deals with the Gender Classification from ECG signal using Least Square Support Vector Machine (LS-SVM) and Support Vector Machine (SVM) Techniques. The different features extracted from ECG signal using Heart Rate Variability (HRV) analysis are the input to the LS-SVM and SVM classifier and at the output the classifier, classifies whether the patient corresponding to required ECG is male or female. The least square formulation of support vector machine (SVM) has been derived from statistical learning theory. SVM has already been marked as a novel development by learning from examples based on polynomial function, neural networks, radial basis function, splines and other functions. The performance of each classifier is decided by classification rate (CR). Our result confirms the classification ability of LS-SVM technique is much better to classify gender from ECG signal analysis in terms of classification rate than SVM.


Introduction
Gender is almost its most salient feature, and gender classification according to ECG is one of the most Challenging problems in person identification in Bio metrics [1]. Co mpared with other research topics in Bio metrics, the academic researches on gender classification is less. In reality, successful gender classification will boost the performance of Pat ient recognition in large Medical database.
Fro m last two decades It was observed that a variety of pred ict ion models have been p rop osed in the mach ine learning that include time series models, regression models, adaptive neuro-fuzzy inference systems (ANFIS), artificial neural network (ANN) models and SVM models [2]. Due to the effectiveness and smoothness of ANN model, it is widely used in various fields like pattern recognition, regression and classificat ion . For classificat ion and non -linear funct ion estimat ion, the recent ly proposed SVM techn ique is an innovative kind of machine learning method introduced by Vapn ik and co -workers [3,4,5]. Th is method is fu rther enhanced by various investigators for different applications like class ificat ion , featu re ext ract ion , clustering , data reduction and regression in different disciplines. SVM have remarkab le generalizat ion performan ce and many mo re adv ant ag es ov er oth er metho ds , an d hence SVM has attracted attention and gained extensive application. Suykens and his group [6] have proposed the use of LS-SVM for simp lification of traditional of SVM. Apart fro m its use in classification in various areas of pattern recognition, it has been extensively used in handling regression problems successfully [7,8]. In LS-SVM , a set of only linear equation (Linear programming) is solved which is much easier and computationally more simple which made it advantageous than SVM .
In the present study, both SVM and LS-SVM classifiers have been designed, trained and tested using various kernel functions like linear and (Radial Basic Function) RBF kernel. RBF kernel LS-SVM gives better performance in terms of classification rate among other classifiers.  In this current study, the ECG of d ifferent people was recorded. An in-house developed ECG data acquisition system was used for this study. The ECG system main ly composed of two parts, viz. ECG electrodes and ECG-DAQ. ECG-DAQ, a USB-based device, helped in recording ECG signals in PC. Figure 1 shows the ECG device connected with PC wh ich was used for data recording in our study.

Materials and Methods
Bio medical Starter Kit fro m National Instrument software was used to extract the different time and frequency domain features of HRV (e.g. mean heart rate (HR), mean RR, Mean NN, standard deviation of RR intervals (SDNN), root mean square of successive difference (RMSSD), NN50, p NN50 and SD1 and SD2 of Poincare plot). The HRV features were then analysed by non-linear statistical analysis using Classification and regression trees (CART) and Boosted tree (BT) classification to determine the significant features in STATISTICA (v7) in figure 2. A co mbination of the features was subsequently used for Gender Classification using SVM and LS-SVM in MATLA B 10.1.
After prediction of significant features like(RR Mean, RR Standard Deviation(std.),HR mean, HR Std, root mean square of successive difference (RMSSD), NN50, pNN50,LF peak, HF Peak, LF Power, HF Po wer, LF/HF ratio) fro m CA RT and BT then these feature values are the input to both SVM classifier. At the output of both classifiers we will classify t wo classes i.e. (boy and girl).the b inary values assigns to girl class as 0 and for boy class the value is 1.After classification we co mpare both the values in terms of Classification rate (CR).
The input data are normalized before being processed in the network as follows: In this method of normalizat ion, the maximu m values of the input vector component are given by:

SVM Classification
SVM technique is an attractive kind of mach ine learn ing method introduced by Vapnik and co-workers [3,4,5]. This method is further modified by various scientists for different applications like classification, feature ext raction, clustering, data reduction and regression in different discip lines of engineering. Our present analysis is based on the classification of b inary class data by employing SVM technique. This method builds a Hyperplane for separation of data into two classes in simple binary classificat ion of linear separable train ing data vector ( 1 2 Where 'w' and 'b' are weight vector normal to Hyperplane and bias value respectively. New test data according to the classification by SVM classifier is assigned to a class according to sign of decision function as: Testing data belongs to class-1(male class) if Test data belongs to class-2(female class) if Is the decision boundary corresponding to Weight vector and bias value for optimal Hyperplane. Support vectors are obtained by maximizing the distance between the closest training point and the corresponding Hyperplane. This can be done by maximizing the marg in defined as 2 M w = same as minimization of Under the methodologies Different number of mathemat ical algorith ms exists for determining the value of weight and bias under the condition (3.5) and (3.6). One of the most efficient method used in SVM is Quadratic Optimizat ion problem. Its solution of the problem involves construction of dual problem with the use of Lagrange multip lier i α which is given as follows: The equation ( Where i x is support vector for each nonzero value of i α . Hence, the classification function for a test data point x is inner product of support vector and test data point, which is given as follows ( ) SVM maps n-dimensional data vector into a d-dimensional feature space (d>n) with help of a mapping function for b inary classification of nonlinear training data points. This Mapping function or kernel function provides a Hyperplane wh ich separate the classes in high dimensional feature space. Using standard Quadratic Programming (QP) optimization technique, the Hyperplane maximizes the margin between classes. Those data point which are Closest to the Hyperplane are used to measure the margin and named as support vectors. In dual formu lation of quadratic optimization problem instead of using dot product of training data points in high dimensional feature space, kernel trick is used. Kernel function defines the inner product of training data points in high dimensional feature space.
Thus the kernel function is defined as The various advantages of kernel function are, It reduces the mathematical as well as the computational co mplexity in higher dimensional feature space. In this paper, co mmon ly used kernel functions are linear, polynomial, radial Gaussian and sigmoid are defined as follows:

LS-SVM Classification
The formulat ion of LS-SVM is introduced as follo ws. ( ) Where w weight is vector and b is the bias term [9,10,11].
As in LS-SVM, it is necessary to minimize a cost function C containing a penalized regression error for binary target, as follows: ( ) The first part of this cost function is a weight decay which is used to regularize weight sizes and penalize large weights. Due to this regularization, the weights converge to similar value. Large weights deteriorate the generalization ability of the LS-SVM because they can cause excessive variance. The second part of eq. (4.2) is the regression error for all training data. The parameter γ , which has to be optimized by the user, gives the relative weight of th is part as compared to the first part. The restriction supplied by eq.
An important result of this approach is that the weights (w) can be written as linear co mbination of the Lagrange mu ltip liers with the corresponding data training (x i ). Putting the result of eq. (4.5) into eq. (4.1), the fo llo wing result is obtained as For a point i y to evaluate it is: The vector follo ws fro m solving a set of linear equations: Where A is a square matrix given by Where K denotes the kernel matrix with ij th element in eq.
(4.5) and I denotes the identity matrix N × N, Hence the solution is given by: All Lagrange mult ipliers (the support vectors) are non-zero, which means that all training objects contribute to the solution and it can be seen fro m eq. (4.10) to eq. (4.11). In contrast with standard SVM the LS-SVM solution is usually not sparse. However, by pruning and reduction techniques a sparse solution can easily achieved.
Depending on the number of train ing data set, either an iterative solver such as conjugate gradients methods (for large data set) can be used or direct solvers, in both cases with nu merically reliable methods.
In application involving nonlinear regression it is not enough to change the inner product of ( ) ( ) Where d is the polynomial degree and 2 sv σ is the squared variance of the Gaussian function, to obtained support vector it should be optimized by user. In order to achieve a good generalization model it is very important to do a careful model selection of the tuning parameters, in co mbination with the regularization constant γ.

Result and Discussion
In this study, the proposed modelling are carried out; 25 sets of input-output patterns used for training both networks and for testing purpose the remaining 12 sets are used. The software programs developed are used for implementation using MATLAB version 10.1. In the beginning, SVM network was trained with co mmon ly used kernels like Linear and RBF Kernel Function. Table (1)(2) shows the performance of model means of Percentage of Classification rates(CR) obtained fro m the processing of testing data with respect to SVM and LS-SVM model with RBF kernel and Linear kernel. Class-2(Girl) 1 5

Conclusions
This paper proposes a gender classification fro m ECG signal using LS-SVM and SVM technique. The different input features extracted fro m HRV analysis are direct ly feed to SVM and LSSVM classifier. Both the classifiers are designed, trained and tested. LS-SVM classifier with RBF kernel g ives better classification rate of 92% among other models. Th is shows LS-SVM as pro mising results for the classification of the Gender based on HRV and ECG signal analysis.