Jaromir Kukal

Wavelet transform use for feature extraction and EEG signal segments classification

Segmentation, feature extraction and classification of signal components belong to very common problems in various engineering, economical and biomedical applications. The paper is devoted to the use of discrete wavelet transform (DWT) both for signal preprocessing and signal segments feature extraction as an alternative to the commonly used discrete Fourier transform (DFT). Feature vectors belonging to separate signal segments are then classified by a competitive neural network as one of methods of cluster analysis and processing. The paper provides a comparison of classification results using different methods of feature extraction most appropriate for EEG signal components detection. Problems of multichannel segmentation are mentioned in this connection as well.

Change in the Characteristics of EEG Color Noise in Alzheimer’s Disease

Neurophysiological experiments support the hypothesis of the presence of critical dynamics of brain activity. This is also manifested by power law of electroencephalography (EEG) power spectra, which can be described by the relation 1/fα. This dependence is a result of internal interactions between parts of the brain and is probably required for optimal processing of information. In Alzheimer’s disease, changes in the functional organization of the brain occur, which may be manifested by changes in the α coefficient. We compared the average values of α for 19 electrodes in the resting EEG record in 110 patients with moderate to severe Alzheimer’s disease (Mini-Mental State Examination [MMSE] score = 10-19) with 110 healthy controls. Statistically, the most significant differences are present in the prefrontal areas. In addition to the prefrontal and frontal areas, the largest separation value in the evaluation of receiver operating characteristic (ROC) curves was recorded in the temporal area. The coefficient alpha has few false-positive results in the optimal operating point of the ROC curve, and is thereby highly specific for Alzheimer’s disease.

Errata: Estimation of non-Gaussian noise parameters in the wavelet domain using the moment-generating function

We discuss methods for modeling and removal of noise in astronomical images. For its favorable properties, we exploit the undecimated wavelet representation and apply noise suppression in this domain. Usually, the noise analysis of the studied imaging system is carried out in the spatial domain. However, noise in astronomical data is non-Gaussian, and thus the noise model parameters need to be estimated directly in the wavelet domain.We derive equations for estimating the sample moments for non-Gaussian noise in the wavelet domain. We consider that the sample moments in the spatial domain are known from the noise analysis and that the model parameters are estimated by using the method of moments.

Cognitive impairment in cerebellar lesions: a logit model based on neuropsychological testing

BackgroundDamage to the cerebellum may lead to motor dysfunctions, but also to the neuropsychological deficits that comprise the Cerebellar Cognitive Affective Syndrome (CCAS). It can affect executive functions, attention, memory, visuospatial functions, language, and emotions. Our goal was to determine which neuropsychological tests could be effectively used to identify this syndrome during a short examination.MethodsTwenty-five patients with an isolated cerebellar lesion and 25 matched healthy controls were examined using an extensive neuropsychological battery.ResultsLogistic regression models and sub-models were computed for individual tests, as well as for the full battery. The best results were produced by a model combining patient education level, the number of errors on the California Verbal Learning Test, and time on Prague Stroop Test (Dots).ConclusionsBased on the results, we suggest that a condensed battery of neuropsychological tests can be used to detect CCAS. The tests are easy to administer and could be helpful in both research and clinical settings.

PAR-2, IL-4R, TGF-β and TNF-α in bronchoalveolar lavage distinguishes extrinsic allergic alveolitis from sarcoidosis.

Sarcoidosis (SARC) and extrinsic allergic alveolitis (EAA) share certain markers, making a differential diagnosis difficult even with histopathological investigation. In lung tissue, proteinase-activated receptor-2 (PAR-2) is primarily investigated with regard to epithelial and inflammatory perspectives. Varying levels of certain chemokines can be a useful tool for distinguishing EAA and SARC. Thus, in the present study, differences in the levels of transforming growth factor (TGF)-β1, tumor necrosis factor (TNF)-α, interleukin-4 receptor (IL-4R) and PAR-2 in bronchoalveolar lavage fluid (BALF) were compared, using an ELISA method, between 14 patients with EAA and six patients with SARC. Statistically significant higher levels of IL-4R, PAR-2 and the PAR-2/TGF-β1 and PAR-2/TNF-α ratios were observed in EAA patients as compared with SARC patients. Furthermore, the ratios of TNF-α/total protein, TGF-β1/PAR-2 and TNF-α/PAR-2 were significantly lower in EAA patients than in SARC patients. The results indicated a higher detection of PAR-2 in EAA samples in association with TNF-α and TGF-β levels. As EAA and PAR-2 in parallel belong to the Th2-mediated pathway, the results significantly indicated an association between this receptor and etiology. In addition, the results indicated that SARC is predominantly a granulomatous inflammatory disease, thus, higher levels of TNF-α are observed. Therefore, the detection of PAR-2 and investigated chemokines in BALF may serve as a useful tool in the differential diagnosis between EAA and SARC.

Higher TGF-β with lower CD124 and TSLP, but no difference in PAR-2 expression in bronchial biopsy of bronchial asthma patients in comparison with COPD patients.

Chronic obstructive pulmonary disease (COPD) and bronchial asthma (BA) are 2 severe respiratory disorders with different predominated immunopathologies. There are several “novel molecules” from different families that are proposed as part of the etiopathogenesis of COPD and BA. Proteinase-activated receptor 2 (PAR-2), thymic stromal lymphoprotein (TSLP), interleukin-4 and its receptor (CD124), Yin-Yang 1 (YY1), and transforming growth factor beta (TGF-&bgr;) have been previously shown to be involved in the pathophysiology of both these diseases. We investigated PAR-2, TSLP, CD124 (interleukin-4R), TGF-&bgr;, and YY1 immunohistochemical expression in endobronchial and transbronchial biopsies from 22 BA patients and 20 COPD patients. Immunostaining for the above-mentioned antigens was quantified using a modified semiquantitative scoring system and statistically evaluated. The values of TGF-&bgr; in the epithelial cells (P=0.0007) and TGF-&bgr; in the submucosa (P=0.0075) were higher in the BA samples, whereas values of CD124 (P=0.0015) and TSLP (P=0.0106) were higher in the COPD samples. No statistically significant differences between the groups were recorded for PAR-2 and YY1. Airway inflammatory reaction diversity in BA and COPD seems to be disease specific; however, there are also shared mechanisms involved in the pathophysiology of both diseases.

Use of fuzzy edge single-photon emission computed tomography analysis in definite Alzheimer's disease - a retrospective study

BackgroundDefinite Alzheimers disease (AD) requires neuropathological confirmation. Single-photon emission computed tomography (SPECT) may enhance diagnostic accuracy, but due to restricted sensitivity and specificity, the role of SPECT is largely limited with regard to this purpose.MethodsWe propose a new method of SPECT data analysis. The method is based on a combination of parietal lobe selection (as regions-of-interest (ROI)), 3D fuzzy edge detection, and 3D watershed transformation. We applied the algorithm to three-dimensional SPECT images of human brains and compared the number of watershed regions inside the ROI between AD patients and controls. The Students two-sample t-test was used for testing domain number equity in both groups.ResultsAD patients had a significantly reduced number of watershed regions compared to controls (p < 0.01). A sensitivity of 94.1% and specificity of 80% was obtained with a threshold value of 57.11 for the watershed domain number. The narrowing of the SPECT analysis to parietal regions leads to a substantial increase in both sensitivity and specificity.ConclusionsOur non-invasive, relatively low-cost, and easy method can contribute to a more precise diagnosis of AD.

Application of optimization heuristics for complex astronomical object model identification

Detection and localization of astronomical objects are two of the most fundamental topics in astronomical science where localization uses detection results. Object localization is based on modeling of point spread function and estimation of its parameters. Commonly used models as Gauss or Moffat in objects localization provide good approximation of analyzed objects but cannot be sufficient in the case of exact applications such as object energy estimation. Thus the use of sophisticated models is upon the place. One of the key roles plays also the way of the objective function estimation. The least square method is often used, but it expects data with normal distribution, thus there is a question of a maximum likelihood method application. Another important factor of presented problem is choice of the right optimization method. Classical methods for objective function minimization usually require a good initial estimate for all parameters and differentiation of the objective function with respect to model parameters. The results indicated that stochastic methods such as simulated annealing or harmony search achieved better results than the classical optimization methods.

Feature selection via competitive levy flights

Evolutionary meta-heuristics are designed for optimization using population with selection and mutation operators. Novelty of our approach is based on competition of various operators from mutation portfolio. Resulting meta-heuristic is successfully tested on the feature selection task: searching for a sparse sub-model having the best possible value by means of information criteria. Beginning with such statistical formulation we obtain an NP-hard optimization task which can be efficiently solved via meta-heuristic approach. We demonstrate how meta-heuristic optimization combined with statistics can enhance machine learning models and therefore is useful in Computational Intelligence in general.

Fast Estimate of Hartley Entropy in Image Sharpening

Two classes of linear IIR filters: Laplacian of Gaussian (LoG) and Difference of Gaussians (DoG) are frequently used as high pass filters for contextual vision and edge detection. They are also used for image sharpening when linearly combined with the original image. Resulting sharpening filters are radially symmetric in spatial and frequency domains. Our approach is based on the radial approximation of unknown optimal filter, which is designed as a weighted sum of Gaussian filters with various radii. The novel filter is designed for MRI image enhancement where the image intensity represents anatomical structure plus additive noise. We prefer the gradient norm of Hartley entropy of whole image intensity as a measure which has to be maximized for the best sharpening. The entropy estimation procedure is as fast as FFT included in the filter but this estimate is a continuous function of enhanced image intensities. Physically motivated heuristic is used for optimum sharpening filter design by its parameter tuning. Our approach is compared with Wiener filter on MRI images.