SEARCH · PubMed Health

Results for “Gaussian process”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 541 records · Page 30Linked to original sources

PubMed2005

Robust time delay estimation of bioelectric signals using least absolute deviation neural network.

The time delay estimation (TDE) is an important issue in modern signal processing and it has found extensive applications in the spatial propagation feature extraction of biomedical signals as well. Due to the extreme complexity and variability of the underlying systems, biomedical signals are usually nonstationary, unstable and even chaotic. Furthermore, due to the limitations of the measurement environments, biomedical signals are often noise-contaminated. Therefore, the TDE of biomedical signals is a challenging issue. A new TDE algorithm based on the least absolute deviation neural network (LADNN) and its application experiments are presented in this paper. The LADNN is the neural implementation of the least absolute deviation (LAD) optimization model, also called unconstrained minimum L1-norm model, with a theoretically proven global convergence. In the proposed LADNN-based TDE algorithm, a given signal is modeled using the moving average (MA) model. The MA parameters are estimated by using the LADNN and the time delay corresponds to the time index at which the MA coefficients have a peak. Due to the excellent features of L1-norm model superior to Lp-norm (p > 1) models in non-Gaussian noise environments or even in chaos, especially for signals that contain sharp transitions (such as biomedical signals with spiky series or motion artifacts) or chaotic dynamic processes, the LADNN-based TDE is more robust than the existing TDE algorithms based on wavelet-domain correlation and those based on higher-order spectra (HOS). Unlike these conventional methods, especially the current state-of-the-art HOS-based TDE, the LADNN-based method is free of the assumption that the signal is non-Gaussian and the noises are Gaussian and, thus, it is more applicable in real situations. Simulation experiments under three different noise environments, Gaussian, non-Gaussian and chaotic, are conducted to compare the proposed TDE method with the existing HOS-based method. Real application experiment is conducted to extract time delay information between every two adjacent channels of gastric myoelectrical activity (GMA) to assess the spatial propagation characteristics of GMA during different phases of the migrating myoelectrical complex (MMC).

Algorithms↗

PubMed2001

Electrophoresis gel image processing and analysis using the KODAK 1D software.

The present article reports on the performance of the KODAK 1D Image Analysis Software for the acquisition of information from electrophoresis experiments and highlights the utility of several mathematical functions for subsequent image processing, analysis, and presentation. Digital images of Coomassie-stained polyacrylamide protein gels containing molecular weight standards and ethidium bromide stained agarose gels containing DNA mass standards are acquired using the KODAK Electrophoresis Documentation and Analysis System 290 (EDAS 290). The KODAK 1D software is used to optimize lane and band identification using features such as isomolecular weight lines. Mathematical functions for mass standard representation are presented, and two methods for estimation of unknown band mass are compared. Given the progressive transition of electrophoresis data acquisition and daily reporting in peer-reviewed journals to digital formats ranging from 8-bit systems such as EDAS 290 to more expensive 16-bit systems, the utility of algorithms such as Gaussian modeling, which can correct geometric aberrations such as clipping due to signal saturation common at lower bit depth levels, is discussed. Finally, image-processing tools that can facilitate image preparation for presentation are demonstrated.

Algorithms↗

PubMed1990

Spectral enhancement to improve the intelligibility of speech in noise for hearing-impaired listeners.

At speech-to-noise ratios between -3 and 6 dB, many hearing-impaired listeners have difficulty in understanding speech, but spectrograms reveal that the formant peaks of voiced speech and some of the spectral peaks associated with unvoiced speech stand out against the background noise. Our speech-enhancement process is based on the assumption that increasing spectral contrast will result in improved intelligibility. The enhancement involves calculating an auditory excitation pattern from the magnitude spectrum of overlapping short segments of the speech signal. This pattern is convolved with a difference-of-Gaussians function whose bandwidth varies with frequency in the same way as the auditory filter bandwidth. Magnitude values from this enhanced pattern are combined with the unchanged phase spectrum from the original signal to produce the enhanced speech. The processing was used to enhance Boothroyd and Bench-Kowal-Bamford Audiometric lists which had been digitally combined with speech-shaped noise at speech-to-noise ratios between -3 and 6 dB. The subjects had moderate to severe sensorineural hearing losses. The processing produced small but significant improvements in intelligibility for the hearing-impaired listeners tested. Possibilities for improving the processing are discussed.

Hearing Loss, Sensorineural↗

PubMed2006

[A robust approach to independent component analysis and its application in the analysis of magnetoencephalographic data].

Independent component analysis (ICA) is a new method of signal statistical processing and widely used in many fields. We face several problems such as the different nature of source signals (e.g. both super-Gaussian and sub-Gaussian sources exist), unknown number of sources and contamination of the sensor signals with a high level of additive noise in the analysis of signal. A robust approach was proposed to solve these problems in this paper. Firstly, observations (noisy data) possessing high dimensionality were preprocessed and decomposed into a source signal subspace and a noise subspace. Then the number of sources was got through the cross-validation method, and this solved the problem that ICA could not confirm the number of sources. At last the transformed low-dimensional source signals were further separated with the fast and stable ICA algorithm. Through the analysis of artificially synthesized data and the real-world Magnetoencephalographic data, the efficacy of this robust approach was illustrated.

Algorithms↗

PubMed2002

The Hermann-Hering grid illusion demonstrates disruption of lateral inhibition processing in diabetes mellitus.

BACKGROUND/AIM: The Hermann-Hering grid illusion consists of dark illusory spots perceived at the intersections of horizontal and vertical white bars viewed against a dark background. The dark spots originate from lateral inhibition processing. This illusion was used to investigate the hypothesis that lateral inhibition may be disrupted in diabetes mellitus. METHOD: A computer monitor based psychophysical test was developed to measure the threshold of perception of the illusion for different bar widths. The contrast threshold for illusion perception at seven bar widths (range 0.09 degrees to 0.60 degrees) was measured using a randomly interleaved double staircase. Convolution of Hermann-Hering grids with difference of Gaussian receptive fields was used to generate model sensitivity functions. The method of least squares was used to fit these to the experimental data. 14 diabetic patients and 12 control subjects of similar ages performed the test. RESULTS: The sensitivity to the illusion was significantly reduced in the diabetic group for bar widths 0.22 degrees, 0.28 degrees, and 0.35 degrees (p = 0.01). The mean centre:surround ratio for the controls was 1:9.1 (SD 1.6) with a mean correlation coefficient of R(2) = 0.80 (SD 0.16). In the diabetic group, two subjects were unable to perceive the illusion. The mean centre:surround ratio for the 12 remaining diabetic patients was 1:8.6 (SD 2.1). However, the correlation coefficients were poor with a mean of R(2) = 0.54 (SD 0.27), p = 0.04 in comparison with the control group. CONCLUSIONS: A difference of Gaussian receptive field model fits the experimental data well for the controls but does not fit the data obtained for the diabetics. This indicates dysfunction of the lateral inhibition processes in the post-receptoral pathway.

Diabetic Retinopathy↗

PubMed2006

Error analysis for rotating active Stokes-Mueller imaging polarimeters.

Retrieval of Mueller matrices from intensity measurements is a noise-sensitive process. In addition, the choice of the method used for extracting Mueller matrix elements greatly influences the precision of the final results. Among available procedures, three have been tested and their robustness analyzed by adding Gaussian noise to computer synthesized data and have been verified by comparing experimental data. As expected, the methods based on classical matrix inversion reach their noise reduction limit even if more data are considered. In contrast, the discrete Fourier transform approach features a remarkable stability. The identified reason is that the retrieval process corresponds to filtering with a periodic kernel.

Journal Article↗

PubMed1988

Standardization and analysis of digitized photographic data in the longitudinal documentation of cataractous growth.

Age-related cataract formation in man can be documented with slit and retroillumination photographs. With digitization and image analysis of such photographs a cataract may be characterized by a frequency distribution of picture elements over a 255 step gray scale spectrum. Transition from a clear to a cataractous lens may be manifested as a change from a unimodal, Gaussian to a multimodal, non Gaussian frequency distribution respectively. How should one compare and contrast these two distributions, so to accurately describe the extent and significance of a change in lens opacification? The in vitro system of cold cataract formation in the rabbit lens was used as a model of the much slower process of age-related cataract formation in man. As in the human lens undergoing progressive opacification, the frequency distribution (number of pixels vs. intensity of gray) for a digitized image of a clear lens at 26 degrees C is unimodal and Gaussian; that of a fully developed cold cataract at 10 degrees C is multimodal and non-gaussian. In spite of the increasing multimodality of the frequency distribution as the temperature dropped and the cataract grew in density and size, the mean gray density proved to be a valid and useful measure to characterize the distribution and to compare different unaligned images. The Wilcoxon Rank Sum Test proved to be useless in comparing the frequency distributions from cataract images because it proved to be too sensitive to subtle changes in the degree of opacification. Anomalous behavior of the opacification process--i.e. clarification as well as opacification of the lens during cold cataract formation caused all pairs to appear statistically significantly different when in appearance there was no difference. The mean of the frequency distribution is less sensitive to this anomalous behavior and is useful as a comparative index. The method of calculating the threshold of significant change in the mean density of a cataract image is presented.

Animals↗

PubMed1999

Apohorseradish peroxidase unfolding and refolding: intrinsic tryptophan fluorescence studies.

The unfolding and refolding of apohorseradish peroxidase, as a function of guanidinium chloride concentration, were monitored by the intrinsic fluorescence intensity, polarization, and lifetime of the single tryptophan residue. The unfolding was reversible and characterized by at least three distinct stages-the intensity and lifetime data, for example, were both characterized by an initial increase followed by a decrease and then a plateau region. The lifetime data, in the absence and presence of guanidinium chloride, were heterogeneous and fit best to a model consisting of a major Gaussian distribution component and a minor, short discrete component. The observed increase in intensity in the initial stage of the unfolding process is attributed to the conversion of this short component into the longer, distributed component as the guanidinium chloride concentration increases. Our results clarify and amplify previous studies on the unfolding of apohorseradish peroxidase by guanidinium chloride.

Apoenzymes↗

PubMed2005

Evidence for new bands in the 3nu1 and 4nu1 regions of propyne.

Vibrationally mediated photodissociation and room-temperature photoacoustic (PA) spectroscopy have been used for obtaining action (monitoring the yield of H photofragments) and absorption spectra of the second (3nu(1)) and third (4nu(1)) C-H acetylenic stretches overtone regions in propyne. The band contours appearing in these regions seem mostly regular even though they are perturbed, as expressed by the origin shifts in different K components, splitting of the K structure, and splitting due to resonances between neighboring states. Symmetric rotor simulations of the band contours of the PA and action spectra allowed extraction of the molecular parameters and rough estimates for the homogeneous broadening arising from energy flow to the bath vibrational states. We particularly benefited from the reduced congestion in the jet-cooled action spectra and their simulations, which enabled observation of yet unknown features in the vicinity of the 3nu(1) and 4nu(1) states. Particularly, the emergence of the new state in the 3nu(1) region was confirmed by the action spectra monitored at several differing jet temperatures, suggesting that it is a dark state in IR vibrational excitation that becomes brighter in UV excitation to the upper electronic state. The monitored and Gaussian-fitted Doppler profiles point to the release of H photofragments with low average translational energies, attributed to an indirect dissociation process occurring after internal conversion to the ground electronic state and isomerization to allene.

Journal Article↗

PubMed2006

Multidimensional infrared spectroscopy of water. I. Vibrational dynamics in two-dimensional IR line shapes.

In this and the following paper, we describe the ultrafast structural fluctuations and rearrangements of the hydrogen bonding network of water using two-dimensional (2D) infrared spectroscopy. 2D IR spectra covering all the relevant time scales of molecular dynamics of the hydrogen bonding network of water were studied for the OH stretching absorption of HOD in D2O. Time-dependent evolution of the 2D IR line shape serves as a spectroscopic observable that tracks how different hydrogen bonding environments interconvert while changes in spectral intensity result from vibrational relaxation and molecular reorientation of the OH dipole. For waiting times up to the vibrational lifetime of 700 fs, changes in the 2D line shape reflect the spectral evolution of OH oscillators induced by hydrogen bond dynamics. These dynamics, characterized through a set of 2D line shape analysis metrics, show a rapid 60 fs decay, an underdamped oscillation on a 130 fs time scale induced by hydrogen bond stretching, and a long time decay constant of 1.4 ps. 2D surfaces for waiting times larger than 700 fs are dominated by the effects of vibrational relaxation and the thermalization of this excess energy by the solvent bath. Our modeling based on fluctuations with Gaussian statistics is able to reproduce the changes in dispersed pump-probe and 2D IR spectra induced by these relaxation processes, but misses the asymmetry resulting from frequency-dependent spectral diffusion. The dynamical origin of this asymmetry is discussed in the companion paper.

Journal Article↗

PubMed2005

Thermodynamics of magnetohydrodynamic flows with axial symmetry.

We present strategies based upon optimization principles in the case of the axisymmetric equations of magnetohydrodynamics (MHD). We derive the equilibrium state by using a minimum energy principle under the constraints of the MHD axisymmetric equations. We also propose a numerical algorithm based on a maximum energy dissipation principle to compute in a consistent way the nonlinearly dynamically stable equilibrium states. Then, we develop the statistical mechanics of such flows and recover the same equilibrium states giving a justification of the minimum energy principle. We find that fluctuations obey a Gaussian shape and we make the link between the conservation of the Casimirs on the coarse-grained scale and the process of energy dissipation. We contrast these results with those of two-dimensional hydrodynamical turbulence where the equilibrium state maximizes a H function at fixed energy and circulation and where the fluctuations are nonuniversal.

Journal Article↗

PubMed2005

Exactly solvable model of continuous stationary 1/f noise.

An exactly solvable model generating a continuous random process with a 1/f power spectrum is presented. Examples of such processes include the angular (phase) speed of trajectories near stable equilibrium points in two-dimensional dynamical systems perturbed by colored Gaussian noise. An exact formula giving the correlation function of the 1/f noise in terms of the correlation of the perturbing colored noises is derived, and used to show that the 1/f spectrum is found in a wide variety of cases. The 1/f noise is non-Gaussian, as demonstrated by calculating its one-time probability distribution function. Numerical simulations confirm and extend the theoretical results.

Journal Article↗

PubMed2000

Automating parallel implementation of neural learning algorithms.

Neural learning algorithms generally involve a number of identical processing units, which are fully or partially connected, and involve an update function, such as a ramp, a sigmoid or a Gaussian function for instance. Some variations also exist, where units can be heterogeneous, or where an alternative update technique is employed, such as a pulse stream generator. Associated with connections are numerical values that must be adjusted using a learning rule, and and dictated by parameters that are learning rule specific, such as momentum, a learning rate, a temperature, amongst others. Usually, neural learning algorithms involve local updates, and a global interaction between units is often discouraged, except in instances where units are fully connected, or involve synchronous updates. In all of these instances, concurrency within a neural algorithm cannot be fully exploited without a suitable implementation strategy. A design scheme is described for translating a neural learning algorithm from inception to implementation on a parallel machine using PVM or MPI libraries, or onto programmable logic such as FPGAs. A designer must first describe the algorithm using a specialised Neural Language, from which a Petri net (PN) model is constructed automatically for verification, and building a performance model. The PN model can be used to study issues such as synchronisation points, resource sharing and concurrency within a learning rule. Specialised constructs are provided to enable a designer to express various aspects of a learning rule, such as the number and connectivity of neural nodes, the interconnection strategies, and information flows required by the learning algorithm. A scheduling and mapping strategy is then used to translate this PN model onto a multiprocessor template. We demonstrate our technique using a Kohonen and backpropagation learning rules, implemented on a loosely coupled workstation cluster, and a dedicated parallel machine, with PVM libraries.

Algorithms↗

PubMed2005

Manipulating saccadic decision-rate distributions in visual search.

The Gaussian shape of reciprocal latency distributions typically found in single saccade tasks supports the idea of a race-to-threshold process underlying the decision when to saccade (R. H. Carpenter & M. L. Williams, 1995). However, second and later saccades in a visual search task revealed decision-rate (=reciprocal latency) distributions that were skewed Gamma-like (E. M. Van Loon, I. T. Hooge, & A. V. Van den Berg, 2002). Here we consider a related family of Beta-prime distributions that follows from strong competition with a signal to stop the sequence, and is described by two parameters: a fixate and saccade threshold. In three saccadic search experiments, we tried to manipulate the two thresholds independently, thereby expecting change in shape and mean of the reciprocal latency distribution. Interestingly, rate distributions for later saccades were significantly better fit by Beta-prime than by Gamma functions. Increases in the distribution's skew were found with higher display density, but only for second and later saccades. First saccade rate distributions were not altered by the expected target location or by visual information presented prior to the search, but making pre-search saccades did influence both thresholds. The mean rate remained a stereotyped function of ordinal position in the saccade sequence. Our results support strong competition between two decision signals underlying the timing of saccades.

Decision Support Techniques↗

PubMed1986

Parallel algorithms for the analysis of two-dimensional electrophoresis gels.

This paper describes some parallel processing algorithms for the analysis of two-dimensional electrophoresis images. The machine used for the processing was the CLIP4 Cellular Array Computer at University College, London, one of the largest processor arrays in the world. Included in this paper are an algorithm for centroid detection, Gaussian fitting algorithms, and an algorithm for the extraction of data out of the cellular array machine. It is shown that these parallel algorithms can run at a speed almost completely independent of the number of spots in the gel images.

Algorithms↗

PubMed1996

Stroop performance in healthy younger and older adults and in individuals with dementia of the Alzheimer's type.

Components of the Stroop task were examined to investigate the role that inhibitory processes play in cognitive changes in healthy older adults and in individuals with dementia of the Alzheimer's type (DAT). Inhibitory breakdowns should result in an increase in Stroop interference. The results indicate that older adults show a disproportionate increase in interference compared with younger adults. DAT individuals show interference proportionate to older adults but a disproportionate increase in facilitation for congruent color-word trials, and an increased intrusion of word naming on incongruent color naming trials. An ex-Gaussian analysis of response time distributions indicated that the increased interference observed in older adults was due to an increase in the tail of the distribution. Application of the process dissociation analysis of the Stroop task (D.S. Lindsay & L.L. Jacoby, 1994) indicated that older adults showed increased word process estimates, whereas DAT individuals showed differences in both color and word process estimates. Taken together, the results are consistent with an inhibitory breakdown in normal aging and an accelerated breakdown in inhibition in DAT individuals.

Adolescent↗

PubMed2004

Distribution of work in isothermal nonequilibrium processes.

Diffusive motion in an externally driven potential is considered. It is shown that the distribution of work required to drive the system from an initial equilibrium state to another is Gaussian for slow but finite driving. Our result is obtained by projection method techniques exploiting a small parameter defined as the switching rate between the two states of the system. The exact solution for a simple model system shows that such an expansion may fail in higher orders, since the mean and the variance following from the exact distribution show nonanalytic behavior.

Journal Article↗

PubMed2004

Gaussian weighting of ocular wave-front measurements.

The measurement of ocular wave-front error gives insight into the optical performance of the eye and possibly a means for assessing visual performance. The visual system responds not only to the quality of the optical image formed on the retina but also to the processing that occurs in the retina and the brain. To develop a metric of visual performance based on wave-front error measurements, these latter processes must somehow be incorporated. In representing the wave-front error in terms of Zernike polynomials, it appears that terms with lower angular frequency have a greater deleterious effect on visual performance than higher-angular-frequency terms. A technique for weighting the pupil function of the eye with a Gaussian filter is demonstrated. It is further demonstrated that the variance of the Gaussian-weighted wave-front error is well correlated with visual performance.

Journal Article↗