Jan Frans Willem Slaets

A new technique to construct a wavelet transform matching a specified signal with applications to digital, real time, spike, and overlap pattern recognition

This work describes a new and different path to create a wavelet transform that can match a specified discrete-time signal. Called Spikelet, it is designed and optimized to spike and overlap pattern recognition in the digitalized signal that comes from H1, a motion-sensitive neuron of the flys visual system. The technique proposed here and the associated algorithm, implemented in real time using a digital signal processor (DSP), are fully detailed. The results obtained matching the signal under analysis show an improvement over all other transforms, including the Daubechies transform. This reassures the efficacy of our transform. rm.

Recording from two neurons: Second-order stimulus reconstruction from spike trains and population coding

We study the reconstruction of visual stimuli from spike trains, representing the reconstructed stimulus by a Volterra series up to second order. We illustrate this procedure in a prominent example of spiking neurons, recording simultaneously from the two H1 neurons located in the lobula plate of the fly Chrysomya megacephala. The fly views two types of stimuli, corresponding to rotational and translational displacements. Second-order reconstructions require the manipulation of potentially very large matrices, which obstructs the use of this approach when there are many neurons. We avoid the computation and inversion of these matrices using a convenient set of basis functions to expand our variables in. This requires approximating the spike train four-point functions by combinations of two-point functions similar to relations, which would be true for gaussian stochastic processes. In our test case, this approximation does not reduce the quality of the reconstruction. The overall contribution to stimulus reconstruction of the second-order kernels, measured by the mean squared error, is only about 5 of the first-order contribution. Yet at specific stimulus-dependent instants, the addition of second-order kernels represents up to 100 improvement, but only for rotational stimuli. We present a perturbative scheme to facilitate the application of our method to weakly correlated neurons.

A complex biological system: the fly's visual module

Is the characterization of biological systems as complex systems in the mathematical sense a fruitful assertion? In this paper we argue in the affirmative, although obviously we do not attempt to confront all the issues raised by this question. We use the flys visual system as an example and analyse our experimental results of one particular neuron in the flys visual system from this point of view. We find that the motion-sensitive ‘H1’ neuron, which converts incoming signals into a sequence of identical pulses or ‘spikes’, encodes the information contained in the stimulus into an alphabet composed of a few letters. This encoding occurs on multilayered sets, one of the features attributed to complex systems. The conversion of intervals between consecutive occurrences of spikes into an alphabet requires us to construct a generating partition. This entails a one-to-one correspondence between sequences of spike intervals and words written in the alphabet. The alphabet dynamics is multifractal both with and without stimulus, though the multifractality increases with the stimulus entropy. This is in sharp contrast to models generating independent spike intervals, such as models using Poisson statistics, whose dynamics is monofractal. We embed the support of the probability measure, which describes the distribution of words written in this alphabet, in a two-dimensional space, whose topology can be reproduced by an M-shaped map. This map has positive Lyapunov exponents, indicating a chaotic-like encoding.

Individual discrimination of freely swimming pulse-type electric fish from electrode array recordings

Abstract Pulse-type weakly electric fishes communicate through electrical discharges with a stereotyped waveform, varying solely the interval between pulses according to the information being transmitted. This simple codification mechanism is similar to the one found in various known neuronal circuits, which renders these animals as good models for the study of natural communication systems, allowing experiments involving behavioral and neuroethological aspects. Performing analysis of data collected from more than one freely swimming fish is a challenge since the detected electric organ discharge (EOD) patterns are dependent on each animal׳s position and orientation relative to the electrodes. However, since each fish emits a characteristic EOD waveform, computational tools can be employed to match each EOD to the respective fish. In this paper we describe a computational method able to recognize fish EODs from dyads using normalized feature vectors obtained by applying Fourier and dual-tree complex wavelet packet transforms. We employ support vector machines as classifiers, and a continuity constraint algorithm allows us to solve issues caused by overlapping EODs and signal saturation. Extensive validation procedures with Gymnotus sp. showed that EODs can be assigned correctly to each fish with only two errors per million discharges.

VSImG: A high frame rate bitmap based display system for neuroscience research

This paper describes a visual stimulus generator (VSImG) capable of displaying a gray-scale, 256x256x8bitmap image with a frame rate of 500Hz using a boustrophedonic scanning technique. It is designed for experiments with motion-sensitive neurons of the flys visual system, where the flicker fusion frequency of the photoreceptors can reach up to 500Hz. Devices with such a high frame rate are not commercially available, but are required, if sensory systems with high flicker fusion frequency are to be studied. The implemented hardware approach gives us complete real-time control of the displacement sequence and provides all the signals needed to drive an electrostatic deflection display. With the use of analog signals, very small high-resolution displacements, not limited by the images pixel size can be obtained. Very slow image displacements with visually imperceptible steps can also be generated. This can be of interest for other vision research experiments. Two different stimulus files can be used simultaneously, allowing the system to generate X-Y displacements on one display or independent movements on two displays as long as they share the same bitmap image.

Introduction to the special issue on Emergent Applications of Fractals and Wavelets in Biology and Biomedicine

It is our great pleasure to edit this special issue on Emergent Applications of Fractals and Wavelets in Biology and Biomedicine, for the Journal of Applied Mathematics and Computation. The topics covered by this special issue are of great interest to the scientific community. We received many submissions and selected eight papers that were critically peer-reviewed ensuring top-quality research and information. A considerable number of new ideas and applications were addressed in these papers, which span a diverse set of methods and applications. The first paper we present in this issue is entitled ‘‘Fractal Binding and Dissociation Kinetics of Prion Proteins on Biosensor Surfaces”, where Reema Taneja and colleagues discuss a fractal analysis for the binding and the dissociation of prion proteins to biosensor surfaces, where both single-and dual-fractal analysis may be used to model the binding and the dissociation kinetics. ‘‘Fractal-based Brain Tumor Detection in Multimodal MRI” is the next paper, where K. M. Iftekharuddin and colleagues investigate the effectiveness of fusing two novel texture features along with intensity in multimodal magnetic resonance (MR) images for pediatric brain tumor segmentation and classification. In the paper ‘‘A Wavelet-based Energetic Approach for the Analysis of Biomedical Signals: applications to the electroencephalogram and electro-oculogram”, Elisa Magosso and colleagues discuss the importance of wavelet transform, which has emerged over recent years as a favoured tool for the investigation of biomedical signals, in order to obtain uniformly distributed atoms of energy across all the scales and to show two different applications of the wavelet-based energetic approach to biomedical signals: a study on epileptic brain electrical activity, and the electro-oculographic tracings aiming at automatic detection of a particular type of eye movement. The next paper of this issue is entitled ‘‘Nonlinear Dynamic Research on EEG Signals in HAI Experiment”, where Wang Xingyuan studies phase space reconstruction techniques and applies them to Electroencephalogram (EEG) signals of piglets resulted from Hypoxic–Asphyxic Injury (HAI) experiments. ‘‘Classification of Normal Swallowing and Oropharyngeal Dysphagia Using Wavelet” is the title of the next paper that is authored by Andre Spadotto and his colleagues. In this work, they present a simple but functional non-invasive method to characterize oropharyngeal dysphagia based on the discrete wavelet transform which is used to analyze the sounds produced when the patient swallows. In the paper ‘‘Construction of a Class of Compactly Supported Symmetric and Balanced Refinable Function Vector By GTST”, Yang Shouzhi introduces the concept of general two-scale similarity transform (GTST) and its applications, giving also two construction examples. ‘‘Fractal Dimension of Time-indexed Paths” is the next paper of this issue, where Attila Imre presents the importance of time-indexing of the track point in order to obtain a precise estimation of the dimension, being the results compared to the traditional divider method. ‘‘Relative Entropy Measures Applied to Healthy and Pathological Voice Characterization” is the last paper, where Paulo Scalassara and colleagues study the characterization of healthy and pathologically affected

Editorial: Introduction to the Special Issue

Welcome to the this special issue of Pattern Recognition Letters, Advances on Pattern Recognition for Speech and Audio Processing. Pattern recognition in speech and audio processing has broken boundaries and provided us with a wide range of important information. Much literature has recently appeared describing innovative ideas in this area, and the topics covered by this special issue are of great interest in the speech and audio, and pattern recognition communities. Particularly, this special issue brings a clear picture of the state-of-the-art in this field. We received many submissions and selected 12 papers that were critically peer-reviewed ensuring top-quality research and information. A considerable number of new ideas and applications are addressed in these papers, that span a diverse set of methods and applications. The first paper we present in this issue is entitled ‘‘A New Look at Discriminative Training for Hidden Markov Models’’. In this paper, Xiaodong He and Li Deng propose a new approach for using the Minimum Classification Error (MCE) criterion to train hidden Markov models (HMMs) for sequential pattern recognition, where a novel optimization method for discriminatively estimating HMM parameters based on growth transformation is provided. In the paper ‘‘Voice Activity Detection based on a Family of Parametric Distributions’’, Jong Won Shin and colleagues introduce the Generalized Gamma Distribution as a new statistical model of spectral distribution to be applied to the likelihood ratio test performed in voice activity detection (VAD). ‘‘Reduction of musical residual noise for speech enhancement using masking properties and optimal smoothing’’ is the third paper of this issue, where Ching-Ta Lu presents a study that aims to reduce the effect of musical residual noise in speech recognition and speech communication systems. In the next article, ‘‘Segmentation of Specific Speech Signals from Multi-dialog Environment Using SVM and Wavelet’’, T.K. Truong and colleagues present a novel multi-speaker segmentation technique, based on wavelets and support vector machines (SVMs). J.E. Rougui and colleagues, in the article ‘‘Organizing Gaussian Mixture Models into a tree for scaling up speaker retrieval’’, address the problem of scaling up speaker recognition to a large number of speakers, by organizing the set of speaker models into a search tree. In the article ‘‘SNR-Dependent Compression of Enhanced Mel Sub-band Energies for Compensation of Noise Effects on MFCC Features’’, Babak Nasersharif and Ahmad Akbari discuss the robustness of Mel-frequency cepstral coefficients (MFCC) in presence of additive noise. In the paper ‘‘Robust Voice Activity Detection Using Perceptual Wavelet-Packet Transform and Teager Energy Operator’’, Shi-Huang Chen and colleagues present a robust voice activity detection (VAD) algorithm that makes use of the perceptual wavelet-packet transform and the Teager energy operator. Waleed Abdulla, in his paper entitled ‘‘Robust Speaker Modeling Using Perceptually Motivated Feature’’, introduces a novel method to extract robust features for text-independent speaker identification from short utterances. Axel Robel and colleagues, in the paper entitled ‘‘On Cepstral and All-Pole based Spectral Envelope Modeling with Unknown Model Order’’, address the problem of investigating the spectral envelope estimation for harmonic speech signals. Marie Roch and colleagues, in the paper ‘‘Foreground auditory scene analysis for hearing aids’’, propose an algorithm to categorize a foreground speaker as opposed to the background noise and parameterize a frequency-based compression algorithm which has been previously shown to improve speech understanding for some individuals with severe sensorineural hearing loss in the 2–3 kHz range. In the paper ‘‘Autoregressive Decomposition and Pole Tracking Applied to Vocal Fold Nodule Signals’’, Paulo Scalassara and his colleagues present a new algorithm to differentiate between normal and pathologically affected voice signals.

Advances on computer-based biological signal processing techniques

It is our great pleasure to edit this special issue, Advances on Computer-based Biological Signal Processing Techniques, for the Computers and Electrical Engineering journal. The topics covered by this special issue are of great interest to the scientific community. We received many submissions and selected eight papers that were critically peer-reviewed ensuring top-quality research and information. A considerable number of new ideas and applications were addressed in these papers, that span a diverse set of methods and applications. The first paper we present in this issue is entitled ‘‘A New Mathematical Based QRS Detector Using Continuous Wavelet Transform’’. In this paper, A. Ghaffari, H. Golbayani, and M. Ghasemi propose a new viewpoint in ECG detection, by using the continuous wavelet transform (CWT), that magnify QRS complex and reduce the effects of other peaks. This is obtained with the novel concept of dominant rescaled wavelet coefficients (DRWC). In the paper ‘‘Studies on the ECG Forward Problem’’, A. Yu and H. Peng introduce several new methods of ECG simulation for the ECG forward problem, dealing with aspects of constructing 3D computer graphics of the heart and the torso, modeling the action potential (AP), and simulating excitation propagation of ventricular myocardium. ‘‘The Hermite transform as an efficient model for local image analysis: An application to medical image fusion’’ is the third paper of this issue, where B. Escalante-Ramı́rez presents the Hermite Transform as an image representation model that can be used to tackle the problem of fusion in multimodal medical imagery. This model includes some important properties of human visual perception, such as local orientation analysis and the Guassian derivative model of early vision. Fusion results are compared with a competitive waveletbased technique, proving that the Hermite transform provides better reconstruction of relevant image structures. In the next article, ‘‘Industrial Application of Machine-in-the-Loop-Learning for Medical Robot Vision System – Concept and Comprehensive Field Study’’, M. Eberhardt, S. Roth and A. Konig discuss the availability of increasingly powerful sensors and processing hardware for vision systems that provides the means for performing more complex signal processing and recognition tasks, and present a conceptual solution to the application of a medical robot vision system for the task of tube type detection. ‘‘A Structural Neural System for Mechanical, Biological, and Environmental Systems’’ is the next paper, where G.R. Kirikera, M.J. Schulz, Y. Yeo-Heung, and V. Shanov discuss the properties of a Structural Neural System (SNS) and continuous sensors that mimic the neurons of the human biological system, that is proposed for monitoring the health of large structures. An example of use of the SNS to monitor damage on a nine meter long wind turbine blade is also described, according to the tests performed at the National Renewable Energy Laboratory, Golden, CO, USA under quasi static loading to determine the strength and ability of the blade to withstand wind loading.

Parallel computing: A case study

Abstract A simple molecular dynamics simulation is used to analyze some speed optimization techniques. The efficiency of sequential and parallel algorithms are discussed. An implementation on a T800 transputer array is proposed and the estimated performance is compared with that obtained on a supercomputer.

Automated pulse discrimination of two freely-swimming weakly electric fish and analysis of their electrical behavior during dominance contest

Electric fishes modulate their electric organ discharges with a remarkable variability. Some patterns can be easily identified, such as pulse rate changes, offs and chirps, which are often associated with important behavioral contexts, including aggression, hiding and mating. However, these behaviors are only observed when at least two fish are freely interacting. Although their electrical pulses can be easily recorded by non-invasive techniques, discriminating the emitter of each pulse is challenging when physically similar fish are allowed to freely move and interact. Here we optimized a custom-made software recently designed to identify the emitter of pulses by using automated chirp detection, adaptive threshold for pulse detection and slightly changing how the recorded signals are integrated. With these optimizations, we performed a quantitative analysis of the statistical changes throughout the dominance contest with respect to Inter Pulse Intervals, Chirps and Offs dyads of freely moving Gymnotus carapo. In all dyads, chirps were signatures of subsequent submission, even when they occurred early in the contest. Although offs were observed in both dominant and submissive fish, they were substantially more frequent in submissive individuals, in agreement with the idea from previous studies that offs are electric cues of submission. In general, after the dominance is established the submissive fish significantly changes its average pulse rate, while the pulse rate of the dominant remained unchanged. Additionally, no chirps or offs were observed when two fish were manually kept in direct physical contact, suggesting that these electric behaviors are not automatic responses to physical contact.