Objective Analysis of Vision Impairments Using Single Trial VEPs

M.Hariharan, Vikneswaran Vijean, Sazali Yaacob School of Mechatronic Engineering

University Malaysia Perlis Perlis, Malaysia

vicky.86max@gmail.com

Abstract— Visually evoked potential (VEP) is an electrical signal generated by the brain (Occipital Cortex) in response to a visual stimuli. These VEP are recorded non-invasively by placing the surface electrodes at the scalp, and observed as a reading on an electroencephalogram (EEG). VEP signal has been widely used for the diagnostics of vision impairments in patients. The main parameters that were considered for the diagnostics of these diseases are the amplitude and the latency values. This field of study is gaining interest from researches all over the world. In this paper, time domain based features of the VEP is studied in an effort to discriminate normal subjects from those having vision impairments. Three different classifiers, Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA) and the k Nearest Neighbor (kNN) are used for the investigation. The proposed method shows promising results in the investigation of vision impairments with accuracy ranging from 69.44% to 100%.

Keywords-visually evoked potential; vision impairement; time domain feature .

I. INTRODUCTION Visually evoked potential is an electrical signal generated

by the brain in response to a visual stimulus and is often used for the diagnostics of ocular diseases. Although VEP is measured from the scalp, it has several distinctive characteristics that differentiate them from the EEG signals [1]. While the VEP corresponds to a specific response of occipital lobe, the EEG signals denote the ongoing activity of wide areas of the cortex [1,2]. Visually evoked potentials are also more sensitive towards the changes in the stimulus conditions compared to the EEG. Patients are usually advised to undergo VEP test by their doctors if they experience any changes in vision that are related to the problems along the visual nerve pathways [2]. The analysis of the VEP signals would help to determine the nature of the diseases. For the past two decades, researchers have studied the responses of VEP in diseases such as past optic neuritis, glaucoma, multiple sclerosis, ocular hypertension, macular degeneration, colour blindness, Parkinson’s disease, idiopathic intracranial hypertension and cataract [1-10]. Traditionally, the amplitude and latency values of the VEP signals have been used for the interpretation of the signals. The amplitudes examined are the N75, P100, and N125 where ‘N’ denotes the negative peak, ‘P’ for positive peak and the numbers (75,100, 125) indicating the time in microseconds. The latency value is measured as the time taken for a visual

stimulus to travel from the eye to the occipital cortex [2,4,11-13]. Ensemble averaging is commonly used to separate the VEP from the background noise to perform the analysis. In ensemble averaging, a minimum of 65 single trials are averaged to obtain the VEP response. This method however, results in information’s related to the variations between the single trials to be lost [14]. Hence, in this research work, the time domain analysis of single trial VEPs in subjects with normal vision and those having vision impairments (near sighted) is performed.

II. VEP RECORDINGS PROCEDURES The standards for conducting VEP experiments have been

established by the international society for clinical electrophysiology of vision (ISCEV). In this regard, the protocols for this study are designed by referring to the ISCEV standards. For the analysis, VEP data is collected from 12 eyes of 6 subjects aged between 24 to 30 years (3 control subjects and 3 subjects with vision impairment and having to wear visual aids). Since the VEP originates from the occipital cortex, each “eye” is treated as an individual “case”. The data is taken from subjects that are seated 60cm in front of a black and white checkerboard pattern that is displayed on a HP L1908w LCD monitor. Two different check sizes, 1 degree and 0.25 degree are used in this study. Subjects are asked to focus their gaze on the red fixation at the centre of the screen while the checkerboard changes phase abruptly at a rate of 2 reversals per second. The monocular VEP recordings are done by placing cup-shaped gold plated electrodes on the scalp overlaying the occipital region. A light tight opaque patch is used for this purpose. The interelectrode resistance was kept below 3kΩ. The active electrodes are placed at Oz and O1 position, while the reference electrode is placed at Fz and the ground electrode is placed on the vertex (Cz position) according to the international 10-20 system shown in Figure 1. The signals are band pass filtered in the range of 0.3 Hz to 100 Hz. The three subjects with vision impairments are asked to wear the visual aids during the recordings.

The experimental data is collected in terms of blocks of trials. One block of trial is the continuous collection of 40 trials displayed one after other. In a typical experiment, 3-4 blocks of trials are recorded. In each block of trials, the eye blink trials are eliminated. Samples are recorded at 1 KHz, for a period of

2012 International Conference on Biomedical Engineering (ICoBE),27-28 February 2012,Penang

978-1-4577-1991-2/12/$26.00 ©2011 IEEE 523

20 seconds. Sixty five artifact free trials of data per eye of the subject are used in this analysis.

Figure 1: Scalp Electrodes placement system according to International 10/20 system [12].

III. PREPROCESSING AND FEATURE EXTRACTION

VEP signals are easily contaminated by noise due to their low amplitudes. Hence, the raw signal must be preprosesed to remove the unwanted information from the signal. The eye blinks are the commonly present artifact in VEP signals. This artifact is removed from the samples by thresholding method in which the trials that are found to have amplitudes more than 100 v is automatically discarded from the experiments using a written program. The treshold value of 100 v is used because the blinking will produce 100-200mV potentials [15]. The single trials are baseline calibrated by removing the mean from the samples.

The preprossesed signals are decomposed into five different frequency bands, delta (0.5Hz-4Hz), theta (4Hz-8Hz), alpha (8Hz-16Hz), beta (16Hz-32Hz) and gamma (32Hz-64Hz) using elliptic digital filter. The filter order is chosen as to give a maximum of 0.1db passband ripple and a minimum of 20db stopband attenuation at ±0.5Hz in respective frequency bands[16]. The filtering is performed twice, forward and reverse filtering as to cancel the phase non-linearity of the elliptic filtering. The statistical time domain features are extracted from the trials for classification. The features used in the analysis are the power, energy, entropy [17], standard deviation (Std) and the kurtosis of the single trial VEP’s and are shown in equation 1-5. The features are computed using a 0.5 second window.

∑ (1) ∑ (2) ∑ log (3) ∑ (4) ∑ (5)

where is the samples, n is the number of samples in a 0.5 second window, is a probability of random phenomena

and is approximated by the difference of spectral components and mean value, | |. The is the mean of .

The extracted features are trained using three different

classifiers, namely k-nearest neighbor (kNN), Linear Discriminant Analysis (LDA) and Quadratic Discriminant Analysis (QDA). The results from the decisive algorithms are validated using 10-fold cross validation method. The whole dataset is randomly divided into 10 mutually exclusive and approximately equal size subsets. The classifiers trains and tests the data for 10 times. In each case, a subset is used as testing data, while the remaining subsets form the training data.

IV. CLASSIFICATION

A. Linear Discriminant Analysis LDA is a well known-scheme for feature classification and

is used in this study for 2-class problem. The LDA classifier reduces the dimensional space while preserving much of the class discriminatory information [18]. This is done by projecting the data onto a lower-dimensional vector space in such a way that the ratio of the inter-class distance and the intra-class distance is maximized, thus achieving maximum discrimination between-class. The maximum likelihood method is used to fit the parameters of this classifier to the available data. The objective of this is to find a linear transformation that gives maximum class separability [19,20].

B. Quadratic Discriminant Analysis A more generalized version of LDA is called the QDA, provided that there are only two classes of points and the measurements are normally distributed. However, in QDA the assumption that the covariance of each class is identical is not taken into consideration. Further, the surface that separates the subspaces will be a conic section (like parabola, hyperbola, etc.) [21].

C. k-nearest neighbor kNN is a simple, supervised algorithm that employs lazy

learning [19,20]. It classifies the test samples based on majority of k-nearest neighbor category. The kNN category is determined by finding the minimum distance between the test samples and each of the training sets. Each of the query instances (test VEP signal) is compared with each of the training instance (training VEP signals). The Euclidean Distance measure is used to find the closest members of the training set to the test class being examined. The label of a class is determined from the kNN category using majority voting. Equation 6 shows the calculation of Euclidean Distance.

∑=

−=N

iiiE babaD

1

2)(),( (6)

where a and b are the training and testing VEP signals composed of N features. The effect of different neighborhood

524

in the classification results is studied by varying the k values from 1 to 4.

D. k-fold cross validation The k-fold cross validation method is used to evaluate and

compare the LDA, QDA and kNN algorithms. Using this method, the data are divided into two segments; one for training and one for testing the classification algorithm. The data are at first divided into k equally sized segments. The k iterations of training and testing are performed in such a way that for every iteration a different segment of data is held-out for validation. The remaining k-1 segments are used for training. This method is used to avoid the overlap between the training and validation sets as they tend to lead to over estimation of the performance of the classifiers. Hence, the results obtained using k-fold cross validation method can provide accurate performance estimation of each algorithm. The value chosen for the analysis is k=10, as it utilizes 90% of

the data for predictions and is more likely to be generalizable for the full data [22].

V. RESULTS AND DISCUSSIONS As to make the computations easier, the features are binary

normalized prior to the classification. The results obtained from the classification for LDA, QDA and kNN classifiers are presented in Table 1. The results for individual check sizes are discussed below.

A. 0.25 degree checks By referring to the Table 1, it can be observed that the

performance of kNN classifier is better than LDA and QDA classifiers. The kNN classifier is able to correctly classify more than 90% of the testing instances for all the five features. From the features tested, Std derived from the time space is found to produce better results for all the classification algorithms. Power, energy and entropy are poorly classified compared to Std and kurtosis by all the three classifiers.

Check Size Classifier Power Energy Entropy2 Std kurtosis

0.25 degree

LDA 74.86 75.47 75.28 91.10 85.36 QDA 69.44 69.08 73.19 98.42 89.28

kNN(1) 93.97 93.97 91.15 99.49 97.44 kNN(2) 94.23 94.49 91.03 99.74 97.18 kNN(3) 95.51 94.62 89.74 99.10 97.31 kNN(4) 94.74 94.49 91.03 99.49 97.31

1 degree

LDA 77.97 78.03 76.86 90.87 77.04 QDA 76.33 76.36 77.04 99.21 94.38

kNN(1) 91.15 90.64 88.97 100.00 98.59 kNN(2) 90.77 90.90 89.10 100.00 98.46 kNN(3) 90.64 90.26 89.10 100.00 97.69 kNN(4) 91.03 92.18 90.26 100.00 98.33

B. 1 degree checks Overall, the same trend is seen in this particular check

size, where the kNN classifier outperforms the LDA and QDA. Perfect classification is achieved with the combination of Std feature and kNN classifier. The repeatability of 100% classification for different nearest neighborhood of kNN shows the reliability of the Std feature.

The classification results for the 1 degree checks are better than the smaller checks. These findings are consistent with the discovery by A. Vincent et al. [23] that the larger checks produce more clear VEP recordings. Although LDA and QDA classifiers performance is low for all the features, the kNN algorithm was able to classify majority of the test samples, resulting in more than 85% accuracy for both the checks. This implies that the non linear classifiers are well

suited for the discrimination of vision impairments than the linear classifiers.

VI. CONCLUSION In this study, the performance of the time domain

features in the classification of vision impairment using single trial VEP is evaluated. The features extracted are tested using LDA, QDA and kNN algorithms. The effect of different k-values in kNN classifiers is also observed in the study. The 100% accuracy obtained from the Std feature for 2 class problem using 1 degree check size reflects the reliability of proposed method. kNN classifier outperforms the LDA and QDA classifier and it shows that the kNN classifier is more suitable for the discrimination problem between the normal subjects and subjects with visual impairments. The time domain approach for single trial VEP is showing promising results and can be extended to the diagnostics of particular ocular diseases. In future, the

TABLE 1: CLASSIFICATIONS RESULTS

525

frequency domain based analysis can also be carried out for the single trial VEP in the effort to develop an automated diagnostics system for ocular diseases.

ACKNOWLEDGMENT

The authors thankfully acknowledge the short-term research grant (No. 9001-00192) from Universiti Malaysia Perlis (UniMAP), Perlis, Malaysia.

REFERENCES [1] S. Sokol, "Visually evoked potentials: theory, techniques and clinical

applications," Survey of ophthalmology, vol. 21, ,1976, pp. 18-44. [2] Visual Evoked Potential (VEP) Summary

http://www.virtualmedicalcentre.com/healthinvest [3] S. L. Graham, A. I. Klistorner, and I. Goldberg, "Clinical application

of objective perimetry using multifocal visual evoked potentials in glaucoma practice," Archives of ophthalmology, vol. 123, 2005, pp. 729-739.

[4] G. Holder, "Electrophysiological assessment of optic nerve disease," Eye, vol. 18,, 2004, pp. 1133-1143.

[5] J. V. Odom, R. Hobson, J. T. Coldren, G. M. Chao, and G. W. Weinstein, "10-Hz flash visual evoked potentials predict post-cataract extraction visual acuity," Documenta ophthalmologica, vol. 66, 1987, pp. 291-299.

[6] Y. H. Shih, Z. J. Huang, and C. E. Chang, "Color pattern-reversal visual evoked potential in eyes with ocular hypertension and primary open-angle glaucoma," Documenta ophthalmologica, vol. 77, 1991, pp. 193-200.

[7] S. Bass, J. Sherman, I. Bodis-Wollner, and S. Nath, "Visual evoked potentials in macular disease," Investigative ophthalmology & visual science, vol. 26, 1985, p. 1071.

[8] A. Kesler, V. Vakhapova, A. D. Korczyn, and V. E. Drory, "Visual evoked potentials in idiopathic intracranial hypertension," Clinical neurology and neurosurgery, vol. 111, 2009, pp. 433-436.

[9] D. Regan and D. Neima, "Visual fatigue and visual evoked potentials in multiple sclerosis, glaucoma, ocular hypertension and Parkinson's disease," Journal of Neurology, Neurosurgery & Psychiatry, vol. 47, 1984, p. 673.

[10] S. Jones, "Visual evoked potentials after optic neuritis," Journal of

neurology, vol. 240, 1993, pp. 489-494. [11] R.Sivakumar , G. R. "Identification of intermediate latencies in

transient visual evoked potentials”, Academic Open Internet Journal, Volume 17, 2006.

[12] J. V. Odom, M. Bach, M. Brigell, G. E. Holder, D. L. McCulloch, and A. P. Tormene, "ISCEV standard for clinical visual evoked potentials (2009 update)," Documenta ophthalmologica, vol. 120, pp. 111-119.

[13] R. Diem, A. Tschirne, and M. Bahr, "Decreased amplitudes in multiple sclerosis patients with normal visual acuity: a VEP study," Journal of clinical neuroscience, vol. 10, pp. 67-70, 2003.

[14] R. Quian Quiroga, "Obtaining single stimulus evoked potentials with wavelet denoising," Physica D: Nonlinear Phenomena, vol. 145, pp. 278-292, 2000.

[15] Palaniappan, R., P. Raveendran, and S. Nishida," Multi-Channel Noise Reduced Visual Evoked Potential Analysis", C ( ) Vol. 123, No. 10, 2003, pp. 1721-1727.

[16] R. Palaniappan, "Vision Related Brain Activity for Biometric Authentication," pp. 3227-3231.

[17] K. Ekštein and T. Pavelka, "Entropy And Entropy-based Features In Signal Processing," 2004.

[18] Farag, A. A. and S. Y. Elhabian “Linear discriminant analysis (LDA). A tutorial on data reduction”, www.cvip.uofl.edu/wwwcvip/education/ECE523/LDA%20Tutorial.pdf [January 2011], 2008.

[19] M.Hariharan, L. S. Chee, O. C. Ai, and S. Yaacob , "Classification of Speech Dysfluencies Using LPC Based Parameterization Techniques", Journal of Medical Systems: 1-10, 2010.

[20] C.H. Park, H. Park, “A comparison of generalized linear discriminant analysis algorithms”, Pattern Recognition, vol. 41, pp. 1083-1097, 2008

[21] S. Bhattacharyya, A. Khasnobish, S. Chatterjee, A. Konar, and D. Tibarewala, "Performance analysis of LDA, QDA and KNN algorithms in left-right limb movement classification from EEG data," pp. 126-131.

[22] P. Refaeilzadeh, L. TANG, and H. LIU, "Cross-Validation," Arizona State University, 2008, pp. 1-6.

[23] A. Vincent, R. Shetty, M. Kurian, and B. K. Shetty, "Prospective, cross-sectional study, demonstrating efficacy of blue fixation target while recording Pattern Visual Evoked Potential in optic neuropathy," Documenta ophthalmologica, vol. 119, pp. 89-99, 2009.

526