Search

Objective: We propose and develop a novel biclustering (N-BiC) approach for performing N-way biclustering of neuroimaging data. Our approach is applicable to an arbitrary number of features from both imaging and behavioral data (e.g., symptoms). We applied it to structural MRI data from patients with schizophrenia. Methods: It uses a source-based morphometry approach [i.e., independent component analysis of gray matter segmentation maps] to decompose the data into a set of spatial maps, each of which includes regions that covary among individuals. Then, the loading parameters for components of interest are entered to an exhaustive search, which incorporates a modified depth-first search technique to carry out the biclustering, with the goal of obtaining submatrices where the selected rows (individuals) show homogeneity in their expressions of selected columns (components) and vice versa. Results: Findings demonstrate that multiple biclusters have an evident association with distinct brain networks for the different types of symptoms in schizophrenia. The study identifies two components: inferior temporal gyrus (16) and brainstem (7), which are related to positive (distortion/excess of normal function) and negative (diminution/loss of normal function) symptoms in schizophrenia, respectively. Conclusion: N-BiC is a data-driven method of biclustering MRI data that can exhaustively explore relationships/substructures from a dataset without any prior information with a higher degree of robustness than earlier biclustering applications. Significance: The use of such approaches is important to investigate the underlying biological substrates of mental illness by grouping patients into homogeneous subjects, as the schizophrenia diagnosis is known to be relatively nonspecific and heterogeneous.

Brain state classification using electroencephalography (EEG) finds applications in both clinical and non-clinical contexts, such as detecting sleep states or perceiving illusory effects during multisensory McGurk paradigm, respectively. Existing literature mostly considers recordings of EEG electrodes that cover the entire head. However, for real world applications, wearable devices that encompass just one (or a few) channels are desirable, which make the classification of EEG states even more challenging. With this as background, we applied variants of data driven Empirical Mode Decomposition (EMD) on McGurk EEG, which is an illusory perception of speech when the movement of lips does not match with the audio signal, for classifying whether the perception is affected by the visual cue or not. After applying a common pre-processing pipeline, we explored four EMD based frameworks to extract EEG features, which were classified using Random Forest. Among the four alternatives, the most effective framework decomposes the ensemble average of two classes of EEG into their respective intrinsic mode functions forming the basis on which the trials were projected to obtain features, which on classification resulted in accuracies of 63.66% using single electrode and 75.85% using three electrodes. The frequency band which plays vital role during audio-visual integration was also studied using traditional band pass filters. Of all, Gamma band was found to be the most prominent followed by alpha and beta bands which contemplates findings from previous studies.

In this paper, we review and discuss brain imaging studies which have used the source-based morphometry (SBM) approach over the past decade. SBM is a data-driven linear multivariate approach for decomposing structural brain imaging data into commonly covarying imaging components and subject-specific loading parameters. It is a well-established technique which has predominantly been used to study neuroanatomic differences between healthy controls and patients with neuropsychiatric diseases. We start by discussing the advantages of this technique over univariate analysis for imaging studies, followed by a discussion of results from recent studies which have successfully applied this methodology. We also present recent extensions of this framework including nonlinear SBM, biclustered independent component analysis (B-ICA) and conclude with the possible directions of work for future.

A novel method for segmentation of heart sounds (HSs) into single cardiac cycle (S"1-Systole-S"2-Diastole) using homomorphic filtering and K-means clustering is presented. Feature vectors were formed after segmentation by using Daubechies-2 wavelet detail coefficients at the second decomposition level. These feature vectors were then used as input to the neural networks. Grow and Learn (GAL) and Multilayer perceptron-Backpropagation (MLP-BP) neural networks were used for classification of three different HSs (Normal, Systolic murmur and Diastolic murmur). It was observed that the classification performance of GAL was similar to MLP-BP. However, the training and testing times of GAL were lower as compared to MLP-BP. The proposed framework could be a potential solution for automatic analysis of HSs that may be implemented in real time for classification of HSs.

Despite its reliable diagnosis, schizophrenia lacks an objective diagnostic test or a validated biomarker, which prevents a better understanding of this disorder. Structural magnetic resonance imaging (sMRI) has been vastly explored to find consistent abnormality patterns of gray matter concentration (GMC) in schizophrenia, yet we are far from having reached conclusive evidence. This paper presents a machine learning approach based on resampling techniques to find brain regions with consistent patterns of GMC differences between healthy controls and schizophrenia patients, these regions being detected by means of source-based morphometry. This work uses multi-site data from the Mind Clinical Imaging Consortium, which is composed of sMRI data from 124 controls and 110 patients. Our method achieves a better classification rate than other algorithms and detects regions with GMC differences between both groups that are consistent with several findings on the literature. In addition, the results obtained on data from multiple sites suggest that it may be possible to replicate these results on other datasets.

For the 24th Machine Learning for Signal Processing competition, participants were asked to automatically diagnose schizophrenia using multimodal features derived from MRI scans. The objective of the classification task was to achieve the best possible schizophrenia diagnosis prediction based only on the multimodal features derived from brain MRI scans. A total of 2087 entries from 291 participants with active Kaggle.com accounts were made. Each participant developed a classifier, with optional feature selection, that combined functional and structural magnetic resonance imaging features. Here we review details about the competition setup, the winning strategies, and provide basic analyses of the submitted entries. We conclude with a discussion of the advances made to the neuroimaging and machine learning fields. Index Terms— Competition, MRI, Schizophrenia, FNC, SBM 1. INTRODUCTION Schizophrenia is a severe and disabling mental illness which has no well-established, non-invasive diagnosis biomarker. Currently, due to its symptom overlap with other mental illnesses (like bipolar disorder) it is diagnosed subjectively only, by process of elimination. In order to broadly explore the current limits of automatic diagnosis of schizophrenia, this year’s machine learning competition was entitled The MLSP 2014 Schizophrenia Classification Challenge, which occurred from June 5 through July 20, 2014. For the second consecutive year, the competition was hosted on Kaggle.com. The response from neuroscience and machine learning communities was fantastic, including 291 participants in 245 teams, and an impressive 2087 valid entries. The competition was sponsored by Kaggle.com, who provided their hosting services, and Springer, who provided monetary rewards to the winning participants. In total, 611 entries from 151 participants outperformed the provided competition benchmark in all performance measures considered here. The three winning strategies used, in ranking order, Gaussian process (GP) [1], Gaussian kernel support vector machine (SVM) [2], and distance weighted discrimination (DWD) [3] classifiers, respectively.

A segmentation algorithm, which detects a single cardiac cycle (S 1-systole-S2-diastole) of phonocardiogram (PCG) signals using homomorphic filtering and K-means clustering and a three way classification of heart sounds into normal (N), systolic murmur (S) and diastolic murmur (D) using grow and learn (GAL) neural network, are presented. Homomorphic filtering converts a non-linear combination of signals (multiplied in time domain) into a linear combination by applying logarithmic transformation. It involves the retrieval of the envelope, a(n) of the PCG signal by attenuating the contribution of fast varying component, f(n) using an appropriate low pass filter. K-means clustering is a non-hierarchical partitioning method, which helps to indicate single cardiac cycle in the PCG signal. Segmentation performance of 90.45% was achieved using the proposed algorithm. Feature vectors were formed after segmentation by using Daubechies-2 wavelet detail coefficients at the second decomposition level. Grow and learn network was used for classification of the segmented PCG signals and a classification accuracy of 97.02% was achieved. It is concluded that homomorphic filtering and GAL network could be used for segmentation and classification of PCG signals without using a reference signal

An algorithm for segmentation of heart sounds (HSs) into a single cardiac cycle (Sl-Systole-S2-Diastole) using homomorphic filtering and k-means clustering and a three way classification of heart sounds into normal (N), systolic murmur (S), and diastolic murmur (D), based on neural networks is developed. This algorithm does not require additional reference signal such as ECG signal. Feature vectors are formed after segmentation by using Daubechies-2 wavelet detail coefficients at the second decomposition level. Redundant features are removed using principal component analysis (PCA). Multilayer perceptron-backpropagation neural network (MLP-BP) is used for classification of three different HSs. A classification accuracy of 94.5% and a segmentation accuracy (or performance) of 90.45% was achieved; thus, demonstrating that segmentation and classification of heart sounds without the aid of reference signal is achievable

This study addresses two important problem statements, namely, selection of training datasets for online Brain-Computer Interface (BCI) classifier training and determination of participant concentration levels during an experiment. The work also attempted a pilot study to integrate electroencephalograms (EEGs) and Near Infra Red Spectroscopy (NIRS) for possible applications such as the BCI and for measuring cognitive levels. Two experiments are presented, the first being a mathematical task interleaved with rest states using NIRS only. In the next, integration of the EEG-NIRS with reference to P300-based BCI systems as well as the experimental conditions designed to elicit the concentration levels (denoted as ON and OFF states here) during the paradigm, are presented. The first experiment indicates that NIRS can be used to differentiate a concentrated (i.e., mental activity) level from the rest. However, the second experiment reveals statistically significant results using the EEG only. We present details about the equipment used, the participants as well as the signal processing and machine learning techniques implemented to analyse the EEG and NIRS data. After discussing the results, we conclude by describing the research scope as well as the possible pitfalls in this work from a NIRS viewpoint, which presents an opportunity for future research exploration for BCI and cognitive performance measures.

Automatic identification of a person’s individuality is an important issue today. Brain Computer Interfaces (BCI) which uses EEG as a modality is a promising area for cognitive biometrics. A BCI system could be used to recognise a sequence (say letters, colours or images) by the user. This sequence could form a ‘BrainWord’, which could be used for authentication in a multimodal environment with other technologies for high security applications. In this work, we studied several variations of the well-known P300 BCI paradigm. The influence of irrelevant stimuli during a task was studied by considering the popular Rapid Serial Visual Paradigm (RSVP) . The variation in spatial locations of the presentation stimuli during a task was studied, by designing a Spatially Varying Paradigm . Comparison of classification accuracies and bit rates for eight participants from a BCI perspective, highlights that RSVP paradigm could be exploited effectively for biometrics.