When considering the probabilistic approach to neural networks in the framework of statistical pattern recognition we assume approximation of class-conditional probability distributions by finite mixtures of product components. The mixture components can be interpreted as probabilistic neurons in neurophysiological terms and, in this respect, the fixed probabilistic description contradicts the well known short-term dynamic properties of biological neurons. By introducing iterative schemes of recognition we show that some parameters of probabilistic neural networks can be "released" for the sake of dynamic processes without disturbing the statistically correct decision making. In particular, we can iteratively adapt the mixture component weights or modify the input pattern in order to facilitate correct recognition. Both procedures are shown to converge monotonically as a special case of the well known EM algorithm for estimating mixtures.

• MeSH
• algoritmy MeSH
• lidé MeSH
• nervová síť MeSH
• neuronové sítě MeSH
• neurony fyziologie   MeSH
• rozpoznávání (psychologie) MeSH
• rozpoznávání automatizované MeSH
• rozpoznávání obrazu fyziologie   MeSH
• statistické modely MeSH
• Check Tag
• lidé MeSH
• Publikační typ
• práce podpořená grantem MeSH

• MeSH
• fantazie fyziologie   MeSH
• lidé MeSH
• modely neurologické MeSH
• počítačová simulace MeSH
• zraková percepce genetika   MeSH
• zrakové dráhy fyziologie   MeSH
• zvířata MeSH
• Check Tag
• lidé MeSH
• zvířata MeSH

Deep learning has recently been utilized with great success in a large number of diverse application domains, such as visual and face recognition, natural language processing, speech recognition, and handwriting identification. Convolutional neural networks, that belong to the deep learning models, are a subtype of artificial neural networks, which are inspired by the complex structure of the human brain and are often used for image classification tasks. One of the biggest challenges in all deep neural networks is the overfitting issue, which happens when the model performs well on the training data, but fails to make accurate predictions for the new data that is fed into the model. Several regularization methods have been introduced to prevent the overfitting problem. In the research presented in this manuscript, the overfitting challenge was tackled by selecting a proper value for the regularization parameter dropout by utilizing a swarm intelligence approach. Notwithstanding that the swarm algorithms have already been successfully applied to this domain, according to the available literature survey, their potential is still not fully investigated. Finding the optimal value of dropout is a challenging and time-consuming task if it is performed manually. Therefore, this research proposes an automated framework based on the hybridized sine cosine algorithm for tackling this major deep learning issue. The first experiment was conducted over four benchmark datasets: MNIST, CIFAR10, Semeion, and UPS, while the second experiment was performed on the brain tumor magnetic resonance imaging classification task. The obtained experimental results are compared to those generated by several similar approaches. The overall experimental results indicate that the proposed method outperforms other state-of-the-art methods included in the comparative analysis in terms of classification error and accuracy.

• MeSH
• algoritmy MeSH
• lidé MeSH
• magnetická rezonanční tomografie MeSH
• nádory mozku * MeSH
• neuronové sítě * MeSH
• psaní rukou MeSH
• Check Tag
• lidé MeSH
• Publikační typ
• časopisecké články MeSH
• práce podpořená grantem MeSH

Současná zdravotnická technika produkuje každým okamžikem velké objemy dat. Výsledkem je informační přetížení a nemožnost zvládnout tato enormní data, např. na odd. intenzivní péče. Nástroje pro vizualizaci dat mají za cíl zmenšit toto informační přetížení pomocí inteligentní abstrakce a vizualizace zajímavých atributů zpracovávaných dat. Nově vyvíjené soft - warové nástroje pro vizualizaci by měly podporovat rychlé porozumění složitým, rozsáhlým a dynamicky rostoucím datovým souborům ve všech oblastech medicíny. Jednou z takových oblastí je analýza a vyhodnocování dlouhodobých záznamů EEG. S vyhodnocováním EEG je spojena celá řada problémů. Jedním z nich je potřeba vizuální kontroly záznamu lékařem. V případě, že lékař musí kontrolovat a hodnotit dlouhodobý záznam EEG, je počítačová podpora analýzy a vizualizace velkou pomocí. Právě možnosti vizualizace EEG záznamů a procesu jejich analýzy jsou předmětem našeho příspěvku.

Healthcare technology produces today large sets of data every second. An information overload results from these enormous data volumes not manageable by physicians, e.g. in intensive care. Data visualization tools aim at reducing the information overload by intelligent abstraction and visualization of the features of interest in the current situation. Newly developed soft - ware tools for visualization should support fast comprehension of complex, large, and dynamically growing datasets in all fi elds of medicine. One of such fi elds is the analysis and evaluation of long–term EEG recordings. One of the problems that are connected with the evaluation of EEG signals is that it necessitates visual checking of such a recording performed by a physician. In case the physician has to check and evaluate long–term EEG recordings computer–aided data analysis and visualization might be of great help. Soft ware tools for visualization of EEG data and data analysis are presented in the paper.

• MeSH
• algoritmy MeSH
• anatomické modely MeSH
• elektroencefalografie využití   MeSH
• epilepsie diagnóza   MeSH
• financování organizované MeSH
• klasifikace MeSH
• kóma diagnóza   patofyziologie   MeSH
• lidé MeSH
• mapování mozku metody   přístrojové vybavení   MeSH
• modely neurologické MeSH
• neuronové sítě MeSH
• počítačové zpracování signálu přístrojové vybavení   MeSH
• shluková analýza MeSH
• spánek fyziologie   MeSH
• zobrazování trojrozměrné MeSH
• Check Tag
• lidé MeSH

Correct assessment of tissue histopathology is a necessary prerequisite for any clinical diagnosis. Nowadays, classical methods of histochemistry and immunohistochemistry are complemented by various techniques adopted from molecular biology and bioanalytical chemistry. Mass spectrometry profiling or imaging offered a new level of tissue visualization in the last decade, revealing hidden patterns of tissue molecular organization. It can be adapted to diagnostic purposes to improve decisions on complex and morphologically not apparent diagnoses. In this work, we successfully combined tissue profiling by mass spectrometry with analysis by artificial neural networks to classify normal and diseased liver and kidney tissues in a mouse model of primary hyperoxaluria type 1. Lack of the liver l-alanine:glyoxylate aminotransferase catalyzing conversion of l-alanine and glyoxylate to pyruvate and glycine causes accumulation of oxalate salts in various tissues, especially urinary system, resulting in compromised renal function and finally end stage renal disease. As the accumulation of oxalate salts alters chemical composition of affected tissues, it makes it available for examination by bioanalytical methods. We demonstrated that the direct tissue MALDI-TOF MS combined with neural computing offers an efficient tool for diagnosis of primary hyperoxaluria type I and potentially for other metabolic disorders altering chemical composition of tissues.

• Klíčová slova
• MALDI-TOF mass spectrometry,
• MeSH
• játra patologie   MeSH
• ledviny patologie   MeSH
• myši MeSH
• neuronové sítě * MeSH
• primární hyperoxalurie * diagnóza   patologie   MeSH
• spektrometrie hmotnostní - ionizace laserem za účasti matrice * statistika a číselné údaje   MeSH
• transaminasy nedostatek   MeSH
• zvířata MeSH
• Check Tag
• myši MeSH
• zvířata MeSH
• Publikační typ
• práce podpořená grantem MeSH

• MeSH
• neuroplasticita MeSH
• poranění nervového systému rehabilitace   MeSH
• regenerace nervu MeSH
• rehabilitace MeSH
• Publikační typ
• monografie MeSH

• Konspekt
• Patologie. Klinická medicína
• NLK Obory
• neurologie
• traumatologie
• neurochirurgie

Angiogenesis is the process of new blood vessels growing from existing vasculature. Visualizing them as a three-dimensional (3D) model is a challenging, yet relevant, task as it would be of great help to researchers, pathologists, and medical doctors. A branching analysis on the 3D model would further facilitate research and diagnostic purposes. In this paper, a pipeline of vision algorithms is elaborated to visualize and analyze blood vessels in 3D from formalin-fixed paraffin-embedded (FFPE) granulation tissue sections with two different staining methods. First, a U-net neural network is used to segment blood vessels from the tissues. Second, image registration is used to align the consecutive images. Coarse registration using an image-intensity optimization technique, followed by finetuning using a neural network based on Spatial Transformers, results in an excellent alignment of images. Lastly, the corresponding segmented masks depicting the blood vessels are aligned and interpolated using the results of the image registration, resulting in a visualized 3D model. Additionally, a skeletonization algorithm is used to analyze the branching characteristics of the 3D vascular model. In summary, computer vision and deep learning is used to reconstruct, visualize and analyze a 3D vascular model from a set of parallel tissue samples. Our technique opens innovative perspectives in the pathophysiological understanding of vascular morphogenesis under different pathophysiological conditions and its potential diagnostic role.

• MeSH
• algoritmy MeSH
• kardiovaskulární fyziologické jevy MeSH
• morfogeneze MeSH
• neuronové sítě * MeSH
• počítačové zpracování obrazu MeSH
• zobrazování trojrozměrné * metody   MeSH
• Publikační typ
• časopisecké články MeSH

Complex spatiotemporal patterns, called chimera states, consist of coexisting coherent and incoherent domains and can be observed in networks of coupled oscillators. The interplay of synchrony and asynchrony in complex brain networks is an important aspect in studies of both the brain function and disease. We analyse the collective dynamics of FitzHugh-Nagumo neurons in complex networks motivated by its potential application to epileptology and epilepsy surgery. We compare two topologies: an empirical structural neural connectivity derived from diffusion-weighted magnetic resonance imaging and a mathematically constructed network with modular fractal connectivity. We analyse the properties of chimeras and partially synchronized states and obtain regions of their stability in the parameter planes. Furthermore, we qualitatively simulate the dynamics of epileptic seizures and study the influence of the removal of nodes on the network synchronizability, which can be useful for applications to epileptic surgery.

• MeSH
• difuzní magnetická rezonance MeSH
• epilepsie diagnostické zobrazování   MeSH
• lidé MeSH
• mozek * fyziologie   MeSH
• nervová síť fyziologie   MeSH
• nervové vedení fyziologie   MeSH
• teoretické modely MeSH
• záchvaty diagnostické zobrazování   MeSH
• Check Tag
• lidé MeSH
• Publikační typ
• práce podpořená grantem MeSH
• srovnávací studie MeSH

Previous studies have demonstrated that humans have a remarkable capacity to memorise a large number of scenes. The research on memorability has shown that memory performance can be predicted by the content of an image. We explored how remembering an image is affected by the image properties within the context of the reference set, including the extent to which it is different from its neighbours (image-space sparseness) and if it belongs to the same category as its neighbours (uniformity). We used a reference set of 2,048 scenes (64 categories), evaluated pairwise scene similarity using deep features from a pretrained convolutional neural network (CNN), and calculated the image-space sparseness and uniformity for each image. We ran three memory experiments, varying the memory workload with experiment length and colour/greyscale presentation. We measured the sensitivity and criterion value changes as a function of image-space sparseness and uniformity. Across all three experiments, we found separate effects of 1) sparseness on memory sensitivity, and 2) uniformity on the recognition criterion. People better remembered (and correctly rejected) images that were more separated from others. People tended to make more false alarms and fewer miss errors in images from categorically uniform portions of the image-space. We propose that both image-space properties affect human decisions when recognising images. Additionally, we found that colour presentation did not yield better memory performance over grayscale images.

• MeSH
• dospělí MeSH
• lidé středního věku MeSH
• lidé MeSH
• neuronové sítě * MeSH
• paměť fyziologie   MeSH
• rozpomínání fyziologie   MeSH
• rozpoznávání (psychologie) fyziologie   MeSH
• rozpoznávání obrazu fyziologie   MeSH
• světelná stimulace metody   MeSH
• Check Tag
• dospělí MeSH
• lidé středního věku MeSH
• lidé MeSH
• mužské pohlaví MeSH
• ženské pohlaví MeSH
• Publikační typ
• časopisecké články MeSH

Spatiotemporal dynamics of event-related potentials (ERP) evoked by non-target stimuli in a visual oddball experiment and the presence of coherent oscillations in beta 2 frequency band of decomposed EEG records from peristimulus period were investigated by means of intracranial electrodes in humans. Twenty-one patients with medically intractable epilepsy participated in the study. The EEG signal was recorded using platinum electrodes implanted in several cortical and subcortical sites. Averaged 2 s EEG records were analyzed. Task-specific EEG changes were found in each patient, ERPs were derived from 92 electrodes used (96 % of possible cases). In the majority of analysed cases, ERPs were composed of several distinct components, and their duration was mostly longer than 1 s. The mean onset of the first ERP component was 158+/-132 ms after the stimulus (median 112 ms, minimum value 42 ms, maximum value 755 ms), and large variability of these onset times was found in all the investigated structures. Possible coherence between neural activities of remote brain sites was investigated by calculating running correlations between pairs of decomposed EEG records (alpha, beta 1, beta 2 frequency bands were used, total number of correlated pairs was 662 in each frequency band). The record pairs exhibiting highly correlated time segments represented 23 % of all the investigated pairs in alpha band, 7 % in beta 1 band, and 59 % in beta 2 band. In investigated 2 s record windows, such segments were distributed evenly, i.e. they were also found before the stimulus onset. In conclusion, the results have implicated the idea that a lot of recorded ERPs was more or less by-products of chance in spreading a signal within the neuronal network, and that their functional relevance was somewhat linked with the phenomenon of activity synchronization.

• MeSH
• beta rytmus EEG MeSH
• časové faktory MeSH
• dospělí MeSH
• epilepsie patofyziologie   psychologie   MeSH
• financování organizované MeSH
• implantované elektrody MeSH
• kognice MeSH
• korová synchronizace MeSH
• lidé středního věku MeSH
• lidé MeSH
• mapování mozku metody   přístrojové vybavení   MeSH
• mozková kůra patofyziologie   MeSH
• nervová síť patofyziologie   MeSH
• periodicita MeSH
• světelná stimulace MeSH
• teorie detekce signálu MeSH
• zraková percepce MeSH
• zrakové dráhy patofyziologie   MeSH
• zrakové evokované potenciály MeSH
• Check Tag
• dospělí MeSH
• lidé středního věku MeSH
• lidé MeSH
• mužské pohlaví MeSH
• ženské pohlaví MeSH