Search results for "pattern recognition"
showing 10 items of 2301 documents
Information-theoretic assessment of cardiovascular-brain networks during sleep
2015
This study was aimed at detecting the structure of the physiological network underlying the regulation of the cardiovascular and brain systems during normal sleep. To this end, we measured from the polysomnographic recordings of 10 healthy subjects the normalized spectral power of heart rate variability in the high frequency band (HF) and the EEG power in the δ, θ, α, σ, and β bands. Then, the causal statistical dependencies within and between these six time series were assessed in terms of internal information (conditional self entropy, CSE) and information transfer (transfer entropy, TE) computed via a linear method exploiting multiple regression models and a nonlinear method combining ne…
An automatic method for metabolic evaluation of gamma knife treatments
2015
Lesion volume delineation of Positron Emission Tomography images is challenging because of the low spatial resolution and high noise level. Aim of this work is the development of an operator independent segmentation method of metabolic images. For this purpose, an algorithm for the biological tumor volume delineation based on random walks on graphs has been used. Twenty-four cerebral tumors are segmented to evaluate the functional follow-up after Gamma Knife radiotherapy treatment. Experimental results show that the segmentation algorithm is accurate and has real-time performance. In addition, it can reflect metabolic changes useful to evaluate radiotherapy response in treated patients.
SIFT Texture Description for Understanding Breast Ultrasound Images
2014
Texture is a powerful cue for describing structures that show a high degree of similarity in their image intensity patterns. This paper describes the use of Self-Invariant Feature Transform (SIFT), both as low-level and high-level descriptors, applied to differentiate the tissues present in breast US images. For the low-level texture descriptors case, SIFT descriptors are extracted from a regular grid. The high-level texture descriptor is build as a Bag-of-Features (BoF) of SIFT descriptors. Experimental results are provided showing the validity of the proposed approach for describing the tissues in breast US images.
Recognition of Cardiac Arrhythmia by Means of Beat Clustering on ECG-Holter Recordings
2012
The follow-up of some cardiac diseases may be achieved by ECG-holter record analysis. A heartbeat clustering method can be used to reduce the usually high computational cost of such Holter analysis. This study describes a method aimed at cardiac arrhythmia recognition based on this approach, by means of unsupervised inspection of morphologically similar heartbeat groups. Singular Value Decomposition (SVD) is used as the feature selection method since the complexity increases exponentially with the number of features. A modification of the k-means algorithm was developed for centroid computation, taking into account heartbeat length changes. Experimental set consisted of ECG records from the…
Measuring Functional Connectivity of Human Intra-Cortex Regions with Total Correlation
2021
The economy of brain organization makes the primate brain consume less energy but efficiency. The neurons densely wired each other dependent on both anatomy structure connectivity and functional connectivity. Here, I only describe functional connectivity with Functional Magnetic Resonance Imaging (fMRI) data. Most importantly, how to quantitative measure information share or separate among functional brain regions, what’s worse, fMRI data exist large dimensional problems or “curse dimensionality” [1]. However, the multivariate total correlation method can perfectly address the above problems. In this paper, two things measured with the information-theoretic technique - total correlation [2,…
Clustering Algorithms for MRI
1991
Magnetic Resonance Imaging (MRI) plays a relevant role in the design of systems for computer assisted diagnosis. MR-images are multi-dimensional in nature; physicians have to combine several perceptual information images to perform the tissue classification needed for diagnosis. Automatic clustering methods help to discriminate relevant features and to perform a preliminary segmentation of the image; it can guide the final manual classification of body-tissues. Three clustering techniques and their integration in a MRI-system are described. Their performance and accuracy was evaluated on synthetic and real image-data. A comparison of our approach with the tissue-classification done by a rad…
Classification of Schizophrenia Patients and Healthy Controls Using ICA of Complex-Valued fMRI Data and Convolutional Neural Networks
2019
Deep learning has contributed greatly to functional magnetic resonance imaging (fMRI) analysis, however, spatial maps derived from fMRI data by independent component analysis (ICA), as promising biomarkers, have rarely been directly used to perform individualized diagnosis. As such, this study proposes a novel framework combining ICA and convolutional neural network (CNN) for classifying schizophrenia patients (SZs) and healthy controls (HCs). ICA is first used to obtain components of interest which have been previously implicated in schizophrenia. Functionally informative slices of these components are then selected and labelled. CNN is finally employed to learn hierarchical diagnostic fea…
Increasing Stability of EEG Components Extraction Using Sparsity Regularized Tensor Decomposition
2018
Tensor decomposition has been widely employed for EEG signal processing in recent years. Constrained and regularized tensor decomposition often attains more meaningful and interpretable results. In this study, we applied sparse nonnegative CANDECOMP/PARAFAC tensor decomposition to ongoing EEG data under naturalistic music stimulus. Interesting temporal, spectral and spatial components highly related with music features were extracted. We explored the ongoing EEG decomposition results and properties in a wide range of sparsity levels, and proposed a paradigm to select reasonable sparsity regularization parameters. The stability of interesting components extraction from fourteen subjects’ dat…
Functional Near Infrared Spectroscopy System Validation for Simultaneous EEG-FNIRS Measurements
2019
Functional near-infrared spectroscopy (fNIRS) applied to brain monitoring has been gaining increasing relevance in the last years due to its not invasive nature and the capability to work in combination with other well–known techniques such as the EEG. The possible use cases span from neural-rehabilitation to early diagnosis of some neural diseases. In this work a wireline FPGA–based fNIRS system, that use SiPM sensors and dual-wavelength LED sources, has been designed and validated to work with a commercial EEG machine without reciprocal interference.
Empirical Mode Decomposition on Mismatch Negativity
2008
Empirical mode decomposition (EMD) has been applied in the various disciplines to extract the desired signal. The basic principle is to decompose a time series into intrinsic mode functions (IFMs) and each IFM corresponds to an oscillation phenomenon. A statistical description of the oscillatory activities of the EEG has been well known. It is desired to extract single oscillatory process from the EEG by EMD. Mismatch negativity (MMN) can be automatically elicited by the deviant stimulus in an oddball paradigm, in which physically the deviant stimulus occurs among repetitive and homogeneous stimuli. MMN thus reflects the ability of the brain to detect changes in auditory stimuli. So, the MM…