Classification of Startle Eyeblink Metrics using Neural Networks
|
|
- Edmund Weaver
- 5 years ago
- Views:
Transcription
1 Classification of Startle Eyeblink Metrics using Neural Networks Christopher T. Lovelace, Reza Derakhshani Member, IEEE, Sriram Pavan Kumar Tankasala Member, IEEE, and Diane L. Filion Abstract-In this paper, we show the feasibility of using high-speed video for measurement of startle eyeblinks as a new augmentative modality for biometric security, as blinks can reveal emotional states of interest in security screenings using nonintrusive measurements. Using neural network as classifiers, this initial study shows that upper eyelid tracking at 250 frames per second can categorize startle blinks with accuracies comparable to those of the well-established but intrusive EMGbased measures of muscles in charge of eyelid closure. Index Terms- Neural Networks, Signal Detection, Biomedical Signal Analysis, Image Processing, Pattern Classification, Biometrics, Psychology. I. INTRODUCTION t is possible to assess a number of psychological processes Iby observing an individual s response to a sudden, intense, and unexpected sensory event. The ensuing startle reaction engages a set of fast, reflexive muscle contractions that includes a robust eyeblink. Numerous studies have shown this startle eyeblink response to be reliably modulated, both in size and latency, by psychological factors such as attention, emotion, cognitive load, and sensory processing [1]. While there are many other approaches to psychological measurement (e.g., questionnaires, functional brain imaging, etc.), startle eyeblink modulation (SEM) has a distinct advantage in that it is a reflexive response. No cognition or overt response is required of the participant, making it difficult to purposely manipulate the test results, which makes SEM a viable modality for security applications. This is especially useful when evaluating whether a person is being deceptive or otherwise in a negative emotional state, arguably the two most important features currently absent in biometric systems which Manuscript received January 15, This work was supported in part by This work was supported in part by a research grant from Center for Identification Research Center (CITeR), a National Science Foundation Industry/University Cooperative Research Center in biometrics. Christopher Lovelace is with the University of Missouri Kansas City, Kansas City, MO USA ( ; lovelacec@ umkc.edu). Reza Derakhshani is with the University of Missouri Kansas City, Kansas City, MO USA, ( derakhshanir@umkc.edu). Sriram Pavan Kumar Tankasala is with the University of Missouri Kansas City, Kansas City, MO USA ( ; Diane Filion is with the University of Missouri Kansas City, Kansas City, MO USA ( ; filiond@ umkc.edu). in large part rely only on recognition of unique, identifiable physiological features such as iris patterns and fingerprints [2]. Two additional factors make SEM useful in detecting imposters or high-risk claimants being vetted by a biometric system. First, as we will show, it is possible to measure SEM unobtrusively using high-speed video. Such recordings are also amenable to long-range acquisition using high-powered optics. Second, the startle response may be modulated by psychological factors related to deception and hostile intent; as it is well-established that features such as startle response intensity and latency, reflected in electromyographic (EMG) recordings, are reliable indicators of emotional valence in response to a preprogrammed stimulus [3]. One can envision a video-based SEM system being used accordingly for humanmachine interactions in future psychometrically enhanced biometric systems. II. MEASUREMENT OF STARTLE EYEBLINK For SEM to be a useful psychological measure, one must be able to quantify the robustness of an eyeblink and its timing relative to the startle-eliciting stimulus. While there are a number of ways to measure the eyeblink response, this is most often done by using contact electrodes to record the EMG activity of the orbicularis occuli muscle just below the eye, whose contraction produces eyelid closure [4]. While the EMG response is a sensitive index of the startle response, this technique has the disadvantage of being fairly obtrusive, requiring cleaning of the skin and adhesion of electrodes. For some applications, such as next generation psychometricallyenhanced biometric security systems, it would be advantageous to have available a non-contact technique where the eyeblink response could be measured without the need for electrodes in order to detect potentially dangerous or deceptive individuals based on their SEM reaction. We are presently developing two non-contact techniques for measuring startle eyeblink. The first uses near infrared light reflected from the surface of the eye as an indicator of lid closure. Eyelid movement produces a measurable change in the amount of reflected near infrared light, with larger blinks producing greater changes. While this technique has revealed that measurement of physical eyelid movement provides nearly as sensitive an index of SEM as EMG [5], it has the disadvantage that, given current instrumentation, the sensor
2 must be positioned very close to the eye. A second noncontact approach to eyeblink measurement uses high-speed video recording. While this has the advantage of allowing for unobtrusive recording at a greater distance, so far the video record has been more difficult to quantify and score. An answer to this challenge is the subject of this paper. Recording of the EMG and reflected light measures both involve digitization of a continuously changing voltage signal. Extraction of the relevant features (e.g., response onset and peak) for individual eyeblink responses from these signals usually involves an initial pass with computer algorithms that programmatically identify the onset and peak of a response. However, due to noise and other considerations, these values must then be verified by eye by the researcher. This can be a time-consuming process, given that there are usually a few dozen trials to be scored for each testing session. Feature extraction for the video signals is much more challenging. One must track not only eyelid movements, but also head movements so that these may be subtracted out of the final eyelid movement vectors. We present here both a method for extracting eyelid movement information from an eyeblink video record as well as a way to quantify SEM in both video and EMG recordings without user intervention using neural networks. procedures were approved by the UMKC Social Sciences Institutional Review Board. B. Stimuli Acoustic stimuli were presented binaurally using Sennheiser HD590 headphones to participants comfortably seated in a sound- and light-attenuating room. Acoustic stimuli and trigger pulses were controlled using the Presentation software program (Neurobehavioral Systems, Inc.; Albany, CA) on a Pentium class desktop computer. The startleeliciting stimulus was a 50 ms, 105 db SPL(A) (a perceptually-weighted decibel scale) burst of white noise (<1 ms rise/fall). The prepulse was a 20 ms, 70 db SPL(A), 1 khz tone (5 ms rise/fall). The onset of the prepulse preceded the onset of the startle stimulus by 120 ms (see Figure 1). Participants were asked to keep their head stationary and their eyes pointed straight ahead at all times. Stimuli were presented in two sets of trials, with each set containing 20 prepulse trials and 20 control trials (pseudorandom order). Individual trials were separated by sec (random, uniform distribution). III. PREPULSE INHIBITION For the purpose of demonstrating the utility of neural networks to the scoring of eyeblinks recorded using highspeed video and EMG, we elicited startle eyeblinks using a standard stimulus paradigm known to produce robust SEM. This paradigm is based on the observation that the startle response to a loud, sudden sound may be reliably modulated by an innocuous sound that occurs just before the loud sound. When the initial sound (prepulse) precedes the startling sound by about 100 ms, this has the effect of speeding eyeblink latency and reducing eyeblink amplitude [6]. This effect, termed prepulse inhibition (PPI), has been used extensively to study sensory processing and cognition in both healthy controls and patients with various psychiatric disorders [7]. PPI is computed by comparing metrics for eyeblinks that follow a startle stimulus that was preceded by a prepulse (prepulse trials) with those that follow a startle stimulus presented alone (control trials). While PPI is (likely) unrelated to detection of deception, this is a robust effect and will help us evaluate the utility of our approach, as the main aim of this study is to establish the utility of neural networks in measuring SEM using high-speed video in lieu of traditional EMG-based methods. IV. METHODS A. Participants Nineteen healthy adult members of the University of Missouri Kansas City community were recruited for this study. All had self-reported normal hearing and no diagnosis of any psychiatric or neurological condition, or a first-degree relative diagnosed with any such condition. Each participant completed informed consent prior to participating, and all Fig. 1. Graphic depiction of the acoustic stimuli in a control trial (top) and prepulse trial (bottom). C. Eyeblink Measurement and Analysis Ocular EMG was recorded using two Ag/Ag-Cl electrodes adhered to the skin just below the left lower eyelid, with a ground electrode attached at the left temple. The EMG data were digitized at 1 khz using the BIOPAC MP-150 system (BIOPAC Systems, Inc.; Goleta, CA) with a gain of 1000 and filter passband of Hz. The raw data were stored and analyzed offline using in-house software. Prior to analysis, EMG waveforms were filtered using 4 th -order Butterworth filters with -3 db cutoff frequencies at 30 and 400 Hz, respectively. Waveforms were additionally smoothed using a 5-point boxcar filter applied 2 6 times in succession. For each trial that contained a visible response and was not contaminated by a pre-stimulus blink, we measured response onset and peak latency relative to stimulus onset and peak amplitude relative to the 50 ms pre-stimulus baseline. High-speed video recordings were made in a subset of 8 participants using a Fastcam 512 PCI camera (Photron USA, Inc.; San Diego, CA) recording at 250 frames per second with a pixel resolution of 256 x 240. The camera was focused tight on the right eye (opposite the side on which EMG recordings were made; see Figure 2 for a sample frame). Video recording on each trial was triggered 100 ms prior to startle stimulus onset (providing a pre-stimulus baseline) and continued for 1000 ms (250 frames). Because camera memory was limited, video recordings were made on only the first 35 trials for each run.
3 Fig. 2. Single frame from a high-speed video record. V. RESULTS A. Participants Of the 19 participants recruited for this study, 4 were excluded because they showed too few trials containing scorable blinks in either the video or EMG. The final sample consisted of 15 participants (11 women) with a mean age of 22.7 years (SD = 4.4 years). High-speed video was recorded from 4 of these participants (4 women), with a mean age of 22.5 years (SD = 1.0 years). B. Feature Extraction For EMG data, which were recorded simultaneously with the high-speed video and for the purpose of validating the efficacy of video methods for differentiating prepulse from control trials, we used a three-dimensional feature vector containing the numerical values for response onset latency (stimulus onset to response onset time difference), response peak latency (stimulus onset to response peak time difference), and response intensity (peak EMG voltage relative to baseline). These measurements have been traditionally used to quantify SEM [4]. We will refer to these onset-peak three dimensional feature sets, which were garnered by visual inspection, as scheme A (please see Figure 3). For high-speed video data we used two types of features. For the first type, the onset-peak features were extracted by visual inspection in a fashion similar to the aforementioned scheme A for EMG, i.e. a triplet of response latency, peak latency, and intensity for each trial (Figure 3). For the second set of features (scheme B), we extracted the displacement of upper eyelid automatically as follows, and used a snapshot of the whole 1-second displacement signal (250 integer values) for each SEM as their feature vector. Biometric applications of the psychometric information contained in startle eyeblinks suggest the need for nonintrusive measurements through video. Thus accurate isolation and tracking of the eyelid movements without any attached markers or additional lighting is desired. Adding to this already challenging image processing task, typical real world applications also require (near) real-time processing. For this purpose, we chose Motus markerless motion capture and image pattern tracking software (Vicon Motion Systems, Denver, CO). We targeted two regions of interest for Motus pattern tracking: a mid upper eyelid point and a reference eye corner area (Figure 4). The reason for the selection of the eye corner as a reference point is the fact that subjects head movements need to be cancelled out by subtracting the displacement of the latter point from the eyelid movements. Figure 5 shows an example of this method, which provides a vertical trace for blink motion free of head movement artifacts. The depicted signals are the y coordinates for the tracked points on the upper eyelid, eye corner reference point, and their difference. The latter yields the video-based SEM signal used for classification in Scheme B, with each feature vector being a 1-second (250 point) eyelid movement signal. Stimulus Onset Peak latency Blink Onset Peak Intensity Fig. 4. Markers are placed on the eye region in high-speed videos for automatic tracking of the eyelid movements. Marker s y pixel count Blink latency Fig. 3. Blink features for Scheme A. Vertical axis represents either EMG intensity or upper eyelid displacement in high-speed video, and horizontal axis depicts time. A triplet of blink latency, peak latency, and peak intensity is used to describe each SEM. Frame number (time, 4ms) Fig. 5. Eyelid movement tracking using high-speed video. Depicted signals from top: upper eyelid location, eye-corner location (reference point), and the relative movement of eyelid (y upper_lid -y eye_corner ). Vertical axis shows markers y coordinate (pixel count). Horizontal axis depicts the corresponding video frame number at 250 FPS (4ms/frame).
4 C. Neural Network Analysis In order to classify the aforementioned blink features into control and prepulse, we examined a variety of different feedforward neural networks as classifiers with bi-valued targets and hyperbolic tangent nonlinearities. For training, we used gradient descent scaled conjugate gradient algorithm [8], which provides a good balance between performance and speed. The SEM feature sets (EMG scheme A and video schemes A and B) were randomly divided into three subsets: 60% for training, 20% for early-stopping/validation, and 20% for testing. Each neural network configuration was initialized by randomizing its weights and subsequently trained 20 times to evaluate the expected performance of each configuration, given the stochastic nature of the local minima convergence of gradient descent with random starting points [9]. The earlystopping maximum fail criterion was set to 5 iterations. All the experiments were performed using MATLAB 2008a software (Mathworks, MA) running on a Pentium-class personal computer. We performed a semi-exhaustive search across different hidden layer sizes in an attempt to canvass the architecture space for best network configurations for our feature sets (EMG scheme A and video schemes A and B). Given the scarcity of training data due to the limited enrollment exposures in typical biometric applications, we chose a regularized error function given in (1) to penalize extraneous weights and thus avoid over-parameterization which is characteristic of smaller training datasets: SSE Regularize d = 1 2 N i= 1 P 2 λ e i + w 2 k = 1 Here N is the total number of training samples, e i is the classification error for i th data point, P is the total number of free parameters (i.e. network weights w k ), and λ is the regularization constant which was set to 0.2 in our experiments. This parameter dictates the weight decay pressure, which is the result of weight-shrinking gradient of the second term in (1). Gradient descent learning lowers the above sum of squared errors (SSE) which creates sparse and thus low variance models on limited training datasets resulting in better generalization capabilities (another justification of the above augmented error criterion comes from derivation of (1) via a Bayesian framework and assuming Gaussian distribution for both the neural network weights and target data noise [10]). Here we chose 2 hidden layer networks especially for scheme B high-speed video features, because not only the addition of another hidden layer will introduce higher computational capabilities and mapping flexibility to the network, but also the first hidden layer can act as a feature map (feature extractor) similar to convolutional networks [11]; explaining the better experimental results garnered using an extra hidden layer. The following are the results of the neural network classifiers, also depicted in Figures 6 through 11. EMG: For the latency-peak feature sets (scheme A), we varied the number of hidden nodes in hidden layer 1 and 2 from 1 to 51, and with a step size of 2. We initialized each possible 2 k (1) combination of the aforementioned layer sizes 20 times by randomizing the weights prior to each training run. The highest achieved average performance (correct classification of unseen prepulse vs. control trials) was 63.7%, which occurred at hidden layer 1 size of 37 and hidden layer 2 size of 39 nodes (Figure 6). At this configuration, and out of the 20 training runs, the maximum attained accuracy was 72.4% (Figure 7). Fig. 6. Average test-set performance of neural network blink classifiers vs. number of nodes in their hidden layer 1 (horizontal, 1-51 nodes) and hidden layer 2 (vertical, 1-51 nodes); used for picking the best architecture. Network inputs are EMG latency-peak triplet features (scheme A).Lighter shades show better test performance. High-speed video, scheme A: For these visually scored triplets features, similar to the previous case, we varied both hidden layer sizes from 1 to 51 nodes with a step size of 2. Using the 20 training runs per configuration, best average performance (60.5%, Figure 8) occurred at layer 1 size of 41 and layer 2 size of 53 nodes. At this configuration, and out of the 20 training runs, the maximum attained accuracy was 100% (Figure 9). High-speed video, scheme B: In this mode we essentially used all the 250 samples that made up each 1-second upper eyelid displacement signals per SEM as the input to our 2 hidden layer neural networks, with no further feature extraction. Again, similar to the previous case, we varied both hidden layer sizes to find a suitable configuration. Given the increased input feature dimensionality (250 vs. 3), and to save computation time, we varied hidden layer 1 size from 1 to 101 nodes, but with a step size of 5. For hidden layer 2, we varied the size from 1 to 22, with a step size of 3. Using the same 20 training runs per configuration, best average performance (54.7%, Figure 10) occurred at a layer 1 size of 31 and layer 2 size of 4 nodes. At this configuration, and out of the 20 training runs, the maximum attained accuracy was 80% (Figure 11). It is interesting to note that the best result, regardless of average best configuration prescreening across all the combinations of layers 1 and 2 sizes was 87%, which occurred at layer 1 and layer 2 sized of 81 and 13 nodes, respectively. These results are summarized in Table 1.
5 Fig. 7. Best test-set performance (out of 20 initializations per configuration) of neural network blink classifiers vs. number of nodes in their hidden layer 1 (horizontal, 1-51 nodes) and hidden layer 2 (vertical, 1-51 nodes). Network inputs are EMG latency-peak triplet features (scheme A). Lighter shades show better test performance. VI. DISCUSSION Using EMG and video recordings, we have shown that eyeblinks measured using high-speed video contain sufficient information to quantify SEM (at least with respect to determining whether or not a startle stimulus was preceded by a prepulse) compared to the more frequently used, but less convenient, EMG measurements. We have also demonstrated that neural networks may be effectively used for direct classification of control vs. PPI SEMs without the need for manual scoring and feature extraction. Fig. 8. Average test-set performance (out of 20 initializations per configuration) of neural network blink classifiers vs. number of nodes in their hidden layer 1 (horizontal nodes) and hidden layer 2 (vertical nodes), used for picking the best architecture. Network inputs are high-speed video latency-peak triplet features (scheme A). Lighter shades show better test performance. Fig. 9. Best test-set performance (out of 20 initializations per configuration) of neural network blink classifiers vs. number of nodes in their hidden layer 1 (horizontal nodes) and hidden layer 2 (vertical, 1-51 nodes), used for picking the best architecture. Network inputs are high-speed video latencypeak triplet features (scheme A). Lighter shades show better test performance. As intimated in the Introduction, a video-based SEM psychometric mechanism could be used, either by itself or in conjunction with next-generation psychometrically-enhanced biometric systems, for noninvasive measurement of psychological constructs relevant to deception. Thus not only the apparent characteristics of the claimants, such as their iris patterns, but also their credibility could be assessed before allowing access to the secured entity protected by a biometric system. While the startle PPI paradigm used in the present study has not yet been specifically tested for sensitivity to deception, it has been suggested elsewhere that other forms of SEM may be useful in this arena [12]. The reason for this is that startle can be reliably modulated by ongoing emotional and attentional states as well as to cognitive load. Therefore, the startle response could have utility in the detection of deception in two ways. First, threatening situations and a state of fear have both been shown to increase the size of the startle response [13]. For example, the startle eyeblink response is larger in the presence of a cue that predicts a painful electric shock [14]. So, to whatever degree being presented with crime-relevant details or deception-related questions induces a state of threat or fear, startle eyeblinks elicited while a person is presented with such stimuli would be assumed to be larger than those presented during innocuous cues. Second, there is evidence to suggest that the act of lying is more cognitively demanding than telling the truth [15] [16], and cognitive effort has been shown to influence PPI. For example, PPI is more robust when a person is actively attending to a prepulse stimulus than when they are not [1]. TABLE I CLASSIFICATION RESULT SUMMARY (TEST DATASET, % CORRECT) Input Type Feature set: Scheme A Feature set: Scheme B EMG Average: 63.7% Best: 72.4% NA High-speed video Average: 60.5% Best: 100% Average: 54.7% Best: 80%
6 ACKNOWLEDGEMENTS The authors wish to extend their gratitude to Wade Elmore and Elizabeth Duval (Department of Psychology, University of Missouri-Kansas City) for their help with participant recruitment, data collection and analysis, as well as Dr. Judee Burgoon and her team (University of Arizona) and Dr. Arun Ross (West Virginia University) for their contributions to this research. REFERENCES Fig. 10. Best Average test-set performance of neural network blink classifiers vs. number of nodes in their hidden layer 1 (horizontal, nodes) and hidden layer 2 (vertical, 1-22 nodes), used for picking the best architecture. Network inputs are high-speed video latency-peak triplet features (scheme A). Lighter shades show better test performance. Increased cognitive load during deception, and/or heightened attention to deception-related cues, may cause a measurable difference in the startle response elicited when a false answer is being formulated and when a truthful answer is about to be given. Future work will include analysis of SEM video captures acquired during deceptive/truthful questions and answers. We also would like to analyze different, smaller feature sets as compared to scheme B in order to be able to utilize smaller data driven classifiers, for instance smaller neural networks, which require smaller datasets enabling more robust training. This should also allow us to calibrate a classifier to each subject, compared to the currently presented results using one classifier for all subjects. Also, while these data were collected with a rather high frame rate of 250 FPS, we would like to find out whether a lower frame rate could be adequate. Fig. 11. Best test-set performance (out of 20 initializations per configuration) of neural network blink classifiers vs. number of nodes in their hidden layer 1 (horizontal, nodes) and hidden layer 2 (vertical, 1-22 nodes), used for picking the best architecture. Network inputs are high-speed video latencypeak triplet features (scheme A). Lighter shades show better test performance. [1] D. L. Filion, M. E. Dawson, and A. M. Schell, "The psychological significance of human startle eyeblink modification: A review," Biological Psychology, vol. 47, pp. 1-43, [2] A. K. Jain, R. Bolle, and S. Pankanti, Eds., BIOMETRICS: Personal Identification in Networked Society. Kluwer Academic Publishers, [3] J. Stanley and R. G. Knight, "Emotional specificity of startle potentiation during the early stages of picture viewing," Psychophysiology, vol. 41, p , [4] T. Blumenthal and et al., "Committee Report: Guidelines for human startle eyeblink electromyographic studies," Psychophysiology, vol. 42, pp. 1-15, [5] C. T. Lovelace, W. R. Elmore, and D. L. Filion, "An alternative to EMG for measuring prepulse inhibition of startle eyeblink," Pyschophysiology, vol. 43, pp , [6] F. K. Graham, "Control of reflex blink excitability," in Neural mechanisms of goal-directed behavior and learning. New York, USA, Academic Press, 1980, pp [7] D. L. Braff, M. A. Geyer, and N. R. Swerdlow, "Human studies of prepulse inhibition of startle: Normal subjects, patient groups, and pharmacological studies," Psychopharmacology, vol. 156, pp , [8] M. F. Møller, A Scaled Conjugate Gradient Algorithm for Fast Supervised Learning. Denmark, Aarhus University Publishing, [9] J. C. Principe, N. R. Euliano, and W. C. Lefebvre, Neural and Adaptive Systems:Fundamentals through Simulations, ISBN ed. New York City, USA, John Wiley and Sons, [10] C. M. Bishop, Pattern Recognition and Machine Learning. New York, USA, Springer Science, [11] S. Haykin, Neural Networks - A Comprehensive Foundation, Second Edition ed. Upper Saddle River, New Jersey, USA, Prentice Hall, [12] B. E. Verschuere, "Startling secrets: Startle eye blink modulation by concealed crime information," Biol Psychol, vol. 76, pp , [13] M. M. Bradley, T. Silakowski, and P. J. Lang, "Fear of pain and defensive action," Pain, vol. 137, pp , [14] A. O. Hamm and E. al., "Emotional learning, hedonic change, and the startle probe," Journal of Abnormal Psychology, vol. 102, pp , [15] R. J. Johnson, "The self in conflict: The role of executive processes during truthful and deceptive responses about attitudes," Neuroimage, vol. 39, pp , [16] S. E. Leal, "the time of the crime: Cognitively induced tonic arousal suppression when lying in a free recall context," Acta APsychol (Amst), vol. 129, pp. 1-7, [17] G. J. Siegel, N. Ichikawa, and S. Steinhauer, "Blink before and after you think: Blinks occur prior to and following cognitive load indexed by pupillary response," Psychophysiology, vol. 45, pp , [18] Discovering Vicon Motus 9 Manual. [19] Vicon Online Help [Online]
7 Christopher T. Lovelace, Ph.D., has been a faculty member in the UMKC Department of Psychology since His research focuses on the use of startle eyeblink and psychophysics to measure sensory and cognitive processing both within and across sense modalities in humans. Dr. Lovelace earned his B.A in Psychology from Wake Forest University, and his Ph.D. in Psychology from American University. He has published research in the areas of startle eyeblink, synesthesia, and multisensory integration. Reza Derakhshani, Ph.D.,(M 98) - joined the UMKC School of Computing and Engineering as an assistant professor in His research focuses on computational intelligence with applications in biometrics and biomedical signal analysis. He earned his Ph.D. and Master s degrees in Computer Engineering and Electrical Engineering respectively from West Virginia University. He earned his bachelor's degree in Electrical Engineering from Iran University of Science and Technology. His work has been mainly funded by the National Science Foundation, and has resulted in a number of peerreviewed publications and a U.S. patent. Sriram Pavan Kumar Tankasala M.S.E.E., (M 08) - is a Ph.D. student at the University of Missouri - Kansas City majoring in Electrical Engineering. His research interests include biometrics and biomedical signal processing. He received his Master s degree in Electrical Engineering from University of Missouri Kansas City, and his bachelor's degree in Electrical and Electronics Engineering from Jawaharlal Nehru Technological University Hyderabad, India. Diane Filion, Ph.D., -is associate professor and chair of the Psychology Department at UMKC. Her research focuses on the use of psychophysiological measures, primarily startle eyeblink, to investigate stimulus, task, and participant factors that improve and impair early attentional processing and sensorimotor gating. Dr. Filion earned her Ph.D. and master s degree from the University of Southern California, and her bachelor s degree from Eastern Washington University.
NUMEROUS advances have been made in the field of
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, VOL. 2, NO. 1, JANUARY-JUNE 2011 1 An Ensemble Method for Classifying Startle Eyeblink Modulation from High-Speed Video Records Reza R. Derakhshani, Member, IEEE,
More informationEBCC Data Analysis Tool (EBCC DAT) Introduction
Instructor: Paul Wolfgang Faculty sponsor: Yuan Shi, Ph.D. Andrey Mavrichev CIS 4339 Project in Computer Science May 7, 2009 Research work was completed in collaboration with Michael Tobia, Kevin L. Brown,
More informationAttention and prepulse inhibition: the effects of task-relevant, irrelevant, and no-task conditions
International Journal of Psychophysiology 56 (2005) 121 128 www.elsevier.com/locate/ijpsycho Attention and prepulse inhibition: the effects of task-relevant, irrelevant, and no-task conditions Gary L.
More informationInformation Processing During Transient Responses in the Crayfish Visual System
Information Processing During Transient Responses in the Crayfish Visual System Christopher J. Rozell, Don. H. Johnson and Raymon M. Glantz Department of Electrical & Computer Engineering Department of
More informationFrequency Tracking: LMS and RLS Applied to Speech Formant Estimation
Aldebaro Klautau - http://speech.ucsd.edu/aldebaro - 2/3/. Page. Frequency Tracking: LMS and RLS Applied to Speech Formant Estimation ) Introduction Several speech processing algorithms assume the signal
More informationAffective reactions to briefly presented pictures
Psychophysiology, 38 ~2001!, 474 478. Cambridge University Press. Printed in the USA. Copyright 2001 Society for Psychophysiological Research Affective reactions to briefly presented pictures MAURIZIO
More informationAccuScreen ABR Screener
AccuScreen ABR Screener Test Methods Doc no. 7-50-1015-EN/02 0459 Copyright notice No part of this Manual or program may be reproduced, stored in a retrieval system, or transmitted, in any form or by any
More informationVIDEO SURVEILLANCE AND BIOMEDICAL IMAGING Research Activities and Technology Transfer at PAVIS
VIDEO SURVEILLANCE AND BIOMEDICAL IMAGING Research Activities and Technology Transfer at PAVIS Samuele Martelli, Alessio Del Bue, Diego Sona, Vittorio Murino Istituto Italiano di Tecnologia (IIT), Genova
More informationEffects of perceptual processing demands on startle eyeblink modification
Psychophysiology, 42 (2005), 440 446. Blackwell Publishing Inc. Printed in the USA. Copyright r 2005 Society for Psychophysiological Research DOI: 10.1111/j.1469-8986.2005.00296.x Effects of perceptual
More informationEffects of perceptual load on startle reflex modification at a long lead interval
Psychophysiology, 43 (2006), 498 503. Blackwell Publishing Inc. Printed in the USA. Copyright r 2006 Society for Psychophysiological Research DOI: 10.1111/j.1469-8986.2006.00420.x Effects of perceptual
More informationPCA Enhanced Kalman Filter for ECG Denoising
IOSR Journal of Electronics & Communication Engineering (IOSR-JECE) ISSN(e) : 2278-1684 ISSN(p) : 2320-334X, PP 06-13 www.iosrjournals.org PCA Enhanced Kalman Filter for ECG Denoising Febina Ikbal 1, Prof.M.Mathurakani
More informationIMPC phenotyping SOPs in JMC
IMPC phenotyping SOPs in JMC Acoustic startle and Pre-pulse Inhibition (PPI) IMPC_ACS_003 Purpose The acoustic startle response is characterized by an exaggerated flinching response to an unexpected strong
More informationAssessment of Reliability of Hamilton-Tompkins Algorithm to ECG Parameter Detection
Proceedings of the 2012 International Conference on Industrial Engineering and Operations Management Istanbul, Turkey, July 3 6, 2012 Assessment of Reliability of Hamilton-Tompkins Algorithm to ECG Parameter
More informationUSING AUDITORY SALIENCY TO UNDERSTAND COMPLEX AUDITORY SCENES
USING AUDITORY SALIENCY TO UNDERSTAND COMPLEX AUDITORY SCENES Varinthira Duangudom and David V Anderson School of Electrical and Computer Engineering, Georgia Institute of Technology Atlanta, GA 30332
More informationA SYSTEM FOR CONTROL AND EVALUATION OF ACOUSTIC STARTLE RESPONSES OF ANIMALS
A SYSTEM FOR CONTROL AND EVALUATION OF ACOUSTIC STARTLE RESPONSES OF ANIMALS Z. Bureš College of Polytechnics, Jihlava Institute of Experimental Medicine, Academy of Sciences of the Czech Republic, Prague
More informationStandard Operating Procedure
1.0 Purpose: 1.1 An acoustic startle model of sensorimotor gating, in which a weak acoustic stimulus (the pre-pulse) is used to decrease the reflex response (startle) produced by a second, more intense,
More informationDevelopment of Ultrasound Based Techniques for Measuring Skeletal Muscle Motion
Development of Ultrasound Based Techniques for Measuring Skeletal Muscle Motion Jason Silver August 26, 2009 Presentation Outline Introduction Thesis Objectives Mathematical Model and Principles Methods
More informationAFOSR PI Meeting Dec 1-3, 2014 Program Director: Dr. Darema Dynamic Integration of Motion and Neural Data to Capture Human Behavior
AFOSR PI Meeting Dec 1-3, 2014 Program Director: Dr. Darema Dynamic Integration of Motion and Neural Data to Capture Human Behavior Dimitris Metaxas(PI, Rutgers) D. Pantazis (co-pi, Head MEG Lab, MIT)
More informationBackground noise decreases both prepulse elicitation and inhibition of acoustic startle blink responding
Biological Psychology 72 (2006) 173 179 www.elsevier.com/locate/biopsycho Background noise decreases both prepulse elicitation and inhibition of acoustic startle blink responding Terry D. Blumenthal *,
More informationA multi-process account of startle modulation during affective perception
Psychophysiology, 43 (2006), 486 497. Blackwell Publishing Inc. Printed in the USA. Copyright r 2006 Society for Psychophysiological Research DOI: 10.1111/j.1469-8986.2006.00412.x A multi-process account
More informationBiceps Activity EMG Pattern Recognition Using Neural Networks
Biceps Activity EMG Pattern Recognition Using eural etworks K. Sundaraj University Malaysia Perlis (UniMAP) School of Mechatronic Engineering 0600 Jejawi - Perlis MALAYSIA kenneth@unimap.edu.my Abstract:
More informationABSTRACT 1. INTRODUCTION 2. ARTIFACT REJECTION ON RAW DATA
AUTOMATIC ARTIFACT REJECTION FOR EEG DATA USING HIGH-ORDER STATISTICS AND INDEPENDENT COMPONENT ANALYSIS A. Delorme, S. Makeig, T. Sejnowski CNL, Salk Institute 11 N. Torrey Pines Road La Jolla, CA 917,
More informationThe functional relationship between visual backward masking and prepulse inhibition
Psychophysiology, 41 (2004), 306 312. Blackwell Publishing Inc. Printed in the USA. Copyright r 2004 Society for Psychophysiological Research DOI: 10.1111/j.1469-8986.2004.00153.x The functional relationship
More informationTake another look: Pupillometry and cognitive processing demands
Take another look: Pupillometry and cognitive processing demands Christine A. Dollaghan Raúl Rojas Jana A. Mueller University of Texas at Dallas Callier Center for Communication Disorders November 19,
More informationREACTION TIME AS A MEASURE OF INTERSENSORY FACILITATION l
Journal oj Experimental Psychology 12, Vol. 63, No. 3, 289-293 REACTION TIME AS A MEASURE OF INTERSENSORY FACILITATION l MAURICE HERSHENSON 2 Brooklyn College In measuring reaction time (RT) to simultaneously
More informationInhibition of acoustic startle using different mechanoreceptive channels
Perception & Psychophysics 2005, 67 (4), 741-747 Inhibition of acoustic startle using different mechanoreceptive channels B. D. HILL and TERRY D. BLUMENTHAL Wake Forest University, Winston-Salem, North
More informationThe impact of numeration on visual attention during a psychophysical task; An ERP study
The impact of numeration on visual attention during a psychophysical task; An ERP study Armita Faghani Jadidi, Raheleh Davoodi, Mohammad Hassan Moradi Department of Biomedical Engineering Amirkabir University
More informationRodent Behavioral Learning and Memory Models. From Mechanisms of Memory, 2 nd Edition by J. David Sweatt, Ph.D.
Rodent Behavioral Learning and Memory Models From Mechanisms of Memory, 2 nd Edition by J. David Sweatt, Ph.D. Hippocampal Pyramidal Neuron of Mice and Rats Figure 1 Open Field Apparatus Open Field Behavior
More informationParametric Optimization and Analysis of Adaptive Equalization Algorithms for Noisy Speech Signals
IOSR Journal of Electrical and Electronics Engineering (IOSR-JEEE) e-issn: 2278-1676, p-issn: 2320 3331, Volume 4, Issue 6 (Mar. -Apr. 2013), PP 69-74 Parametric Optimization and Analysis of Adaptive Equalization
More informationDelineation of QRS-complex, P and T-wave in 12-lead ECG
IJCSNS International Journal of Computer Science and Network Security, VOL.8 No.4, April 2008 185 Delineation of QRS-complex, P and T-wave in 12-lead ECG V.S. Chouhan, S.S. Mehta and N.S. Lingayat Department
More informationDETECTION OF HEART ABNORMALITIES USING LABVIEW
IASET: International Journal of Electronics and Communication Engineering (IJECE) ISSN (P): 2278-9901; ISSN (E): 2278-991X Vol. 5, Issue 4, Jun Jul 2016; 15-22 IASET DETECTION OF HEART ABNORMALITIES USING
More informationECG Beat Recognition using Principal Components Analysis and Artificial Neural Network
International Journal of Electronics Engineering, 3 (1), 2011, pp. 55 58 ECG Beat Recognition using Principal Components Analysis and Artificial Neural Network Amitabh Sharma 1, and Tanushree Sharma 2
More informationIntelligent Edge Detector Based on Multiple Edge Maps. M. Qasim, W.L. Woon, Z. Aung. Technical Report DNA # May 2012
Intelligent Edge Detector Based on Multiple Edge Maps M. Qasim, W.L. Woon, Z. Aung Technical Report DNA #2012-10 May 2012 Data & Network Analytics Research Group (DNA) Computing and Information Science
More informationFIR filter bank design for Audiogram Matching
FIR filter bank design for Audiogram Matching Shobhit Kumar Nema, Mr. Amit Pathak,Professor M.Tech, Digital communication,srist,jabalpur,india, shobhit.nema@gmail.com Dept.of Electronics & communication,srist,jabalpur,india,
More informationNoise Cancellation using Adaptive Filters Algorithms
Noise Cancellation using Adaptive Filters Algorithms Suman, Poonam Beniwal Department of ECE, OITM, Hisar, bhariasuman13@gmail.com Abstract Active Noise Control (ANC) involves an electro acoustic or electromechanical
More informationEPIC-PLUS System for auditory evoked potentials and neuro-otology. PEA - VEMPs ASSR - E-ABR FACIAL NERVE
EPIC-PLUS System for auditory evoked potentials and neuro-otology PEA - VEMPs ASSR - E-ABR FACIAL NERVE EPIC-PLUS System for auditory evoked potentials and neuro-otology Epic-plus, a unique solution. With
More informationMammogram Analysis: Tumor Classification
Mammogram Analysis: Tumor Classification Term Project Report Geethapriya Raghavan geeragh@mail.utexas.edu EE 381K - Multidimensional Digital Signal Processing Spring 2005 Abstract Breast cancer is the
More informationHeart Abnormality Detection Technique using PPG Signal
Heart Abnormality Detection Technique using PPG Signal L.F. Umadi, S.N.A.M. Azam and K.A. Sidek Department of Electrical and Computer Engineering, Faculty of Engineering, International Islamic University
More informationSupplemental Information: Task-specific transfer of perceptual learning across sensory modalities
Supplemental Information: Task-specific transfer of perceptual learning across sensory modalities David P. McGovern, Andrew T. Astle, Sarah L. Clavin and Fiona N. Newell Figure S1: Group-averaged learning
More informationClassification of People using Eye-Blink Based EOG Peak Analysis.
Classification of People using Eye-Blink Based EOG Peak Analysis. Andrews Samraj, Thomas Abraham and Nikos Mastorakis Faculty of Computer Science and Engineering, VIT University, Chennai- 48, India. Technical
More informationRemoving ECG Artifact from the Surface EMG Signal Using Adaptive Subtraction Technique
www.jbpe.org Removing ECG Artifact from the Surface EMG Signal Using Adaptive Subtraction Technique Original 1 Department of Biomedical Engineering, Amirkabir University of technology, Tehran, Iran Abbaspour
More informationISSN: ISO 9001:2008 Certified International Journal of Engineering and Innovative Technology (IJEIT) Volume 2, Issue 10, April 2013
ECG Processing &Arrhythmia Detection: An Attempt M.R. Mhetre 1, Advait Vaishampayan 2, Madhav Raskar 3 Instrumentation Engineering Department 1, 2, 3, Vishwakarma Institute of Technology, Pune, India Abstract
More informationDevelopment of 2-Channel Eeg Device And Analysis Of Brain Wave For Depressed Persons
Development of 2-Channel Eeg Device And Analysis Of Brain Wave For Depressed Persons P.Amsaleka*, Dr.S.Mythili ** * PG Scholar, Applied Electronics, Department of Electronics and Communication, PSNA College
More informationNeuromorphic convolutional recurrent neural network for road safety or safety near the road
Neuromorphic convolutional recurrent neural network for road safety or safety near the road WOO-SUP HAN 1, IL SONG HAN 2 1 ODIGA, London, U.K. 2 Korea Advanced Institute of Science and Technology, Daejeon,
More informationIncorporation of Imaging-Based Functional Assessment Procedures into the DICOM Standard Draft version 0.1 7/27/2011
Incorporation of Imaging-Based Functional Assessment Procedures into the DICOM Standard Draft version 0.1 7/27/2011 I. Purpose Drawing from the profile development of the QIBA-fMRI Technical Committee,
More informationPrepulse inhibition of the startle blink reflex is modulated during a memory task requiring prepulses to be encoded for later report
Prepulse inhibition of the startle blink reflex is modulated during a memory task requiring prepulses to be encoded for later report Author Neumann, David Published 2007 Journal Title International Journal
More informationECHOLAB. Screening and Clinical Modality. User Programmable Protocols. Otoacoustic Emissions System TEOAE - DPOAE - ABR - AABR
ECHOLAB Otoacoustic Emissions System - DPOAE - ABR - AABR Screening and Clinical Modality User Programmable Protocols ECHOLAB Otoacoustic Emissions System - DPOAE - ABR - AABR All of Labat s technology
More informationA neural network model for optimizing vowel recognition by cochlear implant listeners
A neural network model for optimizing vowel recognition by cochlear implant listeners Chung-Hwa Chang, Gary T. Anderson, Member IEEE, and Philipos C. Loizou, Member IEEE Abstract-- Due to the variability
More informationEffect of intensity increment on P300 amplitude
University of South Florida Scholar Commons Graduate Theses and Dissertations Graduate School 2004 Effect of intensity increment on P300 amplitude Tim Skinner University of South Florida Follow this and
More informationConsiderations for designing response quantification procedures in non-traditional psychophysiological applications
Considerations for designing response quantification procedures in non-traditional psychophysiological applications Iyer,A.V. 1,3,Cosand,L.D. 2,3,Courtney,C.G. 2,3, Rizzo, A.A. 3, and Parsons,T.D. 3 1
More informationOn the Use of Brainprints as Passwords
9/24/2015 2015 Global Identity Summit (GIS) 1 On the Use of Brainprints as Passwords Zhanpeng Jin Department of Electrical and Computer Engineering Department of Biomedical Engineering Binghamton University,
More informationThe 29th Fuzzy System Symposium (Osaka, September 9-, 3) Color Feature Maps (BY, RG) Color Saliency Map Input Image (I) Linear Filtering and Gaussian
The 29th Fuzzy System Symposium (Osaka, September 9-, 3) A Fuzzy Inference Method Based on Saliency Map for Prediction Mao Wang, Yoichiro Maeda 2, Yasutake Takahashi Graduate School of Engineering, University
More information1 Introduction. Christopher G. Courtney 1,3, Michael E. Dawson 1, Anne M. Schell 2, and Thomas D. Parsons 3
Affective Computer-Generated Stimulus Exposure: Psychophysiological Support for Increased Elicitation of Negative Emotions in High and Low Fear Subjects Christopher G. Courtney 1,3, Michael E. Dawson 1,
More informationMeasuring Focused Attention Using Fixation Inner-Density
Measuring Focused Attention Using Fixation Inner-Density Wen Liu, Mina Shojaeizadeh, Soussan Djamasbi, Andrew C. Trapp User Experience & Decision Making Research Laboratory, Worcester Polytechnic Institute
More informationA Novel Software Solution to Diagnose the Hearing Disabilities In Human Beings
A Novel Software Solution to Diagnose the Hearing Disabilities In Human Beings Prithvi B S Dept of Computer Science & Engineering SVIT, Bangalore bsprithvi1992@gmail.com Sanjay H S Research Scholar, Jain
More informationSUPPLEMENTARY INFORMATION. Table 1 Patient characteristics Preoperative. language testing
Categorical Speech Representation in the Human Superior Temporal Gyrus Edward F. Chang, Jochem W. Rieger, Keith D. Johnson, Mitchel S. Berger, Nicholas M. Barbaro, Robert T. Knight SUPPLEMENTARY INFORMATION
More informationAversive picture processing: Effects of a concurrent task on sustained defensive system engagement
Psychophysiology, 48 (2011), 112 116. Wiley Periodicals, Inc. Printed in the USA. Copyright r 2010 Society for Psychophysiological Research DOI: 10.1111/j.1469-8986.2010.01041.x BRIEF REPORT Aversive picture
More informationSpikerBox Neural Engineering Workshop
SpikerBox Neural Engineering Workshop A Workshop Curriculum for Grades 9-12 Developed as a two-day, two hours/day workshop Developed by UW Graduate Students: Stephanie Seeman, Bethany Kondiles, and Katherine
More informationNeurophysiologically Driven Image Triage: A Pilot Study
Neurophysiologically Driven Image Triage: A Pilot Study Santosh Mathan Honeywell Laboratories 3660 Technology Dr Minneapolis, MN 55418 USA santosh.mathan@honeywell.com Stephen Whitlow Honeywell Laboratories
More informationABR PTA ASSR Multi-ASSR OAE TEOAE DPOAE SOAE VEMP ECochG MLR P300
ABR PTA ASSR Multi-ASSR OAE TEOAE DPOAE SOAE VEMP ECochG MLR P300 Neuro-Audio one device for all audiological tests Auditory brainstem response (ABR)/Brainstem evoked response audiometry (BERA) (air and
More informationThe effect of stimulus modality and task difficulty on attentional modulation of blink startle
The effect of stimulus modality and task difficulty on attentional modulation of blink startle Author Neumann, David, V. Lipp, Ottmar, McHugh, Meredith Published 2004 Journal Title Psychophysiology DOI
More informationArtificial Neural Networks in Cardiology - ECG Wave Analysis and Diagnosis Using Backpropagation Neural Networks
Artificial Neural Networks in Cardiology - ECG Wave Analysis and Diagnosis Using Backpropagation Neural Networks 1.Syed Khursheed ul Hasnain C Eng MIEE National University of Sciences & Technology, Pakistan
More informationClassification of EEG signals in an Object Recognition task
Classification of EEG signals in an Object Recognition task Iacob D. Rus, Paul Marc, Mihaela Dinsoreanu, Rodica Potolea Technical University of Cluj-Napoca Cluj-Napoca, Romania 1 rus_iacob23@yahoo.com,
More informationLocal Image Structures and Optic Flow Estimation
Local Image Structures and Optic Flow Estimation Sinan KALKAN 1, Dirk Calow 2, Florentin Wörgötter 1, Markus Lappe 2 and Norbert Krüger 3 1 Computational Neuroscience, Uni. of Stirling, Scotland; {sinan,worgott}@cn.stir.ac.uk
More informationCHAPTER 6 INTERFERENCE CANCELLATION IN EEG SIGNAL
116 CHAPTER 6 INTERFERENCE CANCELLATION IN EEG SIGNAL 6.1 INTRODUCTION Electrical impulses generated by nerve firings in the brain pass through the head and represent the electroencephalogram (EEG). Electrical
More informationNeuro-Audio Version 2010
ABR PTA ASSR Multi-ASSR OAE TEOAE DPOAE SOAE ECochG MLR P300 Neuro-Audio Version 2010 one device for all audiological tests Auditory brainstem response (ABR)/Brainstem evoked response audiometry (BERA)
More informationSupporting Information
Supporting Information ten Oever and Sack 10.1073/pnas.1517519112 SI Materials and Methods Experiment 1. Participants. A total of 20 participants (9 male; age range 18 32 y; mean age 25 y) participated
More informationPupillary Response Based Cognitive Workload Measurement under Luminance Changes
Pupillary Response Based Cognitive Workload Measurement under Luminance Changes Jie Xu, Yang Wang, Fang Chen, Eric Choi National ICT Australia 1 University of New South Wales jie.jackxu@gmail.com, {yang.wang,
More informationRunning head: HEARING-AIDS INDUCE PLASTICITY IN THE AUDITORY SYSTEM 1
Running head: HEARING-AIDS INDUCE PLASTICITY IN THE AUDITORY SYSTEM 1 Hearing-aids Induce Plasticity in the Auditory System: Perspectives From Three Research Designs and Personal Speculations About the
More informationValence-arousal evaluation using physiological signals in an emotion recall paradigm. CHANEL, Guillaume, ANSARI ASL, Karim, PUN, Thierry.
Proceedings Chapter Valence-arousal evaluation using physiological signals in an emotion recall paradigm CHANEL, Guillaume, ANSARI ASL, Karim, PUN, Thierry Abstract The work presented in this paper aims
More informationAutomatic Detection of Heart Disease Using Discreet Wavelet Transform and Artificial Neural Network
e-issn: 2349-9745 p-issn: 2393-8161 Scientific Journal Impact Factor (SJIF): 1.711 International Journal of Modern Trends in Engineering and Research www.ijmter.com Automatic Detection of Heart Disease
More informationNeural correlates of perceptual separation-induced enhancement of prepulse inhibition of startle in humans
www.nature.com/scientificreports Received: 4 July 2017 Accepted: 4 December 2017 Published: xx xx xxxx OPEN Neural correlates of perceptual separation-induced enhancement of prepulse inhibition of startle
More informationA Brain Computer Interface System For Auto Piloting Wheelchair
A Brain Computer Interface System For Auto Piloting Wheelchair Reshmi G, N. Kumaravel & M. Sasikala Centre for Medical Electronics, Dept. of Electronics and Communication Engineering, College of Engineering,
More informationClassification of Pre-Stimulus EEG of K-complexes using Competitive Learning Networks
Classification of Pre-Stimulus EEG of K-complexes using Competitive Learning Networks Martin Golz 1, David Sommer 1, Thomas Lembcke 2, Brigitte Kurella 2 1 Fachhochschule Schmalkalden, Germany 2 Wilhelm-Griesinger-Krankenhaus,
More informationSparse Coding in Sparse Winner Networks
Sparse Coding in Sparse Winner Networks Janusz A. Starzyk 1, Yinyin Liu 1, David Vogel 2 1 School of Electrical Engineering & Computer Science Ohio University, Athens, OH 45701 {starzyk, yliu}@bobcat.ent.ohiou.edu
More informationAutomatic and controlled attentional processes in startle eyeblink modification: Effects of habituation of the prepulse
Psychophysiology, 37 ~2000!, 409 417. Cambridge University Press. Printed in the USA. Copyright 2000 Society for Psychophysiological Research Automatic and controlled attentional processes in startle eyeblink
More informationThe Simon Effect as a Function of Temporal Overlap between Relevant and Irrelevant
University of North Florida UNF Digital Commons All Volumes (2001-2008) The Osprey Journal of Ideas and Inquiry 2008 The Simon Effect as a Function of Temporal Overlap between Relevant and Irrelevant Leslie
More informationApplication of Artificial Neural Networks in Classification of Autism Diagnosis Based on Gene Expression Signatures
Application of Artificial Neural Networks in Classification of Autism Diagnosis Based on Gene Expression Signatures 1 2 3 4 5 Kathleen T Quach Department of Neuroscience University of California, San Diego
More informationDCNV RESEARCH TOOL: INVESTIGATION OF ANTICIPATORY BRAIN POTENTIALS
DCNV RESEARCH TOOL: INVESTIGATION OF ANTICIPATORY BRAIN POTENTIALS Roman Golubovski, Dipl-Eng J.Sandanski 116-3/24 91000 Skopje, Macedonia Phone: + 389 91 165 367, Fax: + 389 91 165 304 email: roman.golubovski@iname.com
More informationSum of Neurally Distinct Stimulus- and Task-Related Components.
SUPPLEMENTARY MATERIAL for Cardoso et al. 22 The Neuroimaging Signal is a Linear Sum of Neurally Distinct Stimulus- and Task-Related Components. : Appendix: Homogeneous Linear ( Null ) and Modified Linear
More informationA HMM-based Pre-training Approach for Sequential Data
A HMM-based Pre-training Approach for Sequential Data Luca Pasa 1, Alberto Testolin 2, Alessandro Sperduti 1 1- Department of Mathematics 2- Department of Developmental Psychology and Socialisation University
More informationGender Based Emotion Recognition using Speech Signals: A Review
50 Gender Based Emotion Recognition using Speech Signals: A Review Parvinder Kaur 1, Mandeep Kaur 2 1 Department of Electronics and Communication Engineering, Punjabi University, Patiala, India 2 Department
More informationNeuro-MEP-Micro EMG EP. 2-Channel Portable EMG and NCS System with a Built-in Miniature Dedicated Keyboard. EMG according to international standards
Neuro-MEP-Micro 2-Channel Portable EMG and NCS System with a Built-in Miniature Dedicated Keyboard EMG according to international standards Instant analysis of high-quality responses Over 50 EMG and EP
More informationNoise-Robust Speech Recognition in a Car Environment Based on the Acoustic Features of Car Interior Noise
4 Special Issue Speech-Based Interfaces in Vehicles Research Report Noise-Robust Speech Recognition in a Car Environment Based on the Acoustic Features of Car Interior Noise Hiroyuki Hoshino Abstract This
More informationElectro-Oculography I Laboratory
Introduction Sight is probably the most important of the five senses to many human beings. Our visual system continuously provides feedback on objects and interactions with our environment in almost everything
More informationOn the feasibility of speckle reduction in echocardiography using strain compounding
Title On the feasibility of speckle reduction in echocardiography using strain compounding Author(s) Guo, Y; Lee, W Citation The 2014 IEEE International Ultrasonics Symposium (IUS 2014), Chicago, IL.,
More informationQuick Guide - eabr with Eclipse
What is eabr? Quick Guide - eabr with Eclipse An electrical Auditory Brainstem Response (eabr) is a measurement of the ABR using an electrical stimulus. Instead of a traditional acoustic stimulus the cochlear
More informationECG Signal Analysis for Abnormality Detection in the Heart beat
GRD Journals- Global Research and Development Journal for Engineering Volume 1 Issue 10 September 2016 ISSN: 2455-5703 ECG Signal Analysis for Abnormality Detection in the Heart beat Vedprakash Gujiri
More informationBinaural Hearing. Why two ears? Definitions
Binaural Hearing Why two ears? Locating sounds in space: acuity is poorer than in vision by up to two orders of magnitude, but extends in all directions. Role in alerting and orienting? Separating sound
More informationEmotion Classification along Valence Axis Using Averaged ERP Signals
Emotion Classification along Valence Axis Using Averaged ERP Signals [1] Mandeep Singh, [2] Mooninder Singh, [3] Ankita Sandel [1, 2, 3]Department of Electrical & Instrumentation Engineering, Thapar University,
More informationAdvanced Audio Interface for Phonetic Speech. Recognition in a High Noise Environment
DISTRIBUTION STATEMENT A Approved for Public Release Distribution Unlimited Advanced Audio Interface for Phonetic Speech Recognition in a High Noise Environment SBIR 99.1 TOPIC AF99-1Q3 PHASE I SUMMARY
More informationStudy of perceptual balance for binaural dichotic presentation
Paper No. 556 Proceedings of 20 th International Congress on Acoustics, ICA 2010 23-27 August 2010, Sydney, Australia Study of perceptual balance for binaural dichotic presentation Pandurangarao N. Kulkarni
More informationNATIONAL COMPETENCY SKILL STANDARDS FOR PERFORMING NERVE CONDUCTION STUDIES
NATIONAL COMPETENCY SKILL STANDARDS FOR PERFORMING NERVE CONDUCTION STUDIES Nerve Conduction Study (NCS) providers practice in accordance with the facility policy and procedure manual which details every
More informationBiologically-Inspired Human Motion Detection
Biologically-Inspired Human Motion Detection Vijay Laxmi, J. N. Carter and R. I. Damper Image, Speech and Intelligent Systems (ISIS) Research Group Department of Electronics and Computer Science University
More informationNoise-Robust Speech Recognition Technologies in Mobile Environments
Noise-Robust Speech Recognition echnologies in Mobile Environments Mobile environments are highly influenced by ambient noise, which may cause a significant deterioration of speech recognition performance.
More informationDISCRETE WAVELET PACKET TRANSFORM FOR ELECTROENCEPHALOGRAM- BASED EMOTION RECOGNITION IN THE VALENCE-AROUSAL SPACE
DISCRETE WAVELET PACKET TRANSFORM FOR ELECTROENCEPHALOGRAM- BASED EMOTION RECOGNITION IN THE VALENCE-AROUSAL SPACE Farzana Kabir Ahmad*and Oyenuga Wasiu Olakunle Computational Intelligence Research Cluster,
More informationAn electrocardiogram (ECG) is a recording of the electricity of the heart. Analysis of ECG
Introduction An electrocardiogram (ECG) is a recording of the electricity of the heart. Analysis of ECG data can give important information about the health of the heart and can help physicians to diagnose
More informationSpike Sorting and Behavioral analysis software
Spike Sorting and Behavioral analysis software Ajinkya Kokate Department of Computational Science University of California, San Diego La Jolla, CA 92092 akokate@ucsd.edu December 14, 2012 Abstract In this
More informationAn ECG Beat Classification Using Adaptive Neuro- Fuzzy Inference System
An ECG Beat Classification Using Adaptive Neuro- Fuzzy Inference System Pramod R. Bokde Department of Electronics Engineering, Priyadarshini Bhagwati College of Engineering, Nagpur, India Abstract Electrocardiography
More informationPHYSIOLOGICAL RESEARCH
DOMAIN STUDIES PHYSIOLOGICAL RESEARCH In order to understand the current landscape of psychophysiological evaluation methods, we conducted a survey of academic literature. We explored several different
More information