Ramasamy Vijayalakshmi

FP-GraphMiner-A Fast Frequent Pattern Mining Algorithm for Network Graphs

In recent years, graph representations have been used extensively for modelling complicated structural information, such as circuits, images, molecular structures, biological networks, weblogs, XML documents and so on. As a result, frequent subgraph mining has become an important subfield of graph mining. This paper presents a novel Frequent Pattern Graph Mining algorithm, FP-GraphMiner, that compactly represents a set of network graphs as a Frequent Pattern Graph (or FP-Graph). This graph can be used to efficiently mine frequent subgraphs including maximal frequent subgraphs and maximum common subgraphs. The algorithm is space and time efficient requiring just one scan of the graph database for the construction of the FP-Graph, and the search space is significantly reduced by clustering the subgraphs based on their frequency of occurrence. A series of experiments performed on sparse, dense and complete graph data sets and a comparison with MARGIN, gSpan and FSMA using real time network data sets confirm the efficiency of the proposed FP-GraphMiner algorithm.

Minimum connected component - A novel approach to detection of cognitive load induced changes in functional brain networks

Recent advances in computational neuroscience have enabled trans-disciplinary researchers to address challenging tasks such as the identification and characterization of cognitive function in the brain. The application of graph theory has contributed to the modelling and understanding the brain dynamics. This paper presents a new approach based on a special graph theoretic concept called minimum connected component (MCC) to detect cognitive load induced changes in functional brain networks using EEG data. The results presented in this paper clearly demonstrate that the MCC based analysis of the functional brain networks derived from multi-channel EEG data is able to detect and quantify changes across the scalp in response to specific cognitive tasks. The MCC, due to its sensitivity to cognitive load, has the potential to be used as a tool not only to measure cognitive activity quantitatively, but also to detect cognitive impairment.

Computational Techniques for Characterizing Cognition Using EEG Data – New Approaches☆

Abstract Identifying the integrative aspects of brain structure and function, specifically how the connections and interactions among neuronal elements (neurons, brain regions) result in cognition and behavior, is one of the last great frontiers for scientific research. Unraveling the activity of the brains billions of neurons and how they combine to form functional networks has been and remains restricted by both technological and ethical constraints; thus, researchers are increasingly turning to sophisticated data search techniques such as complex network clustering and graph mining algorithms to further delve into the hidden workings of the human mind. By combining such techniques with more traditional inferential statistics and then applying these to multichannel Electroencephalography (EEG) data, it is believed that it is possible to both identify and accurately describe hidden patterns and correlations in functional brain networks, which would otherwise remain undetected. The current paper presents an overview of the application of such approaches to EEG data, bringing together a variety of techniques, including complex network analysis, coherence, mutual information, approximate entropy, computer visualization, signal processing and multivariate techniques such as the one-way analysis of variance (ANOVA). This study demonstrates that the integration of these techniques enables a depth of understanding of complex brain dynamics that is not possible by other methods as well as allowing the identification of differences in system complexity that are believed to underscore normal human cognition.

Transfer Entropy and Information Flow Patterns in Functional Brain Networks during Cognitive Activity

Most previous studies of functional brain networks have been conducted on undirected networks despite the direction of information flow able to provide additional information on how one brain region influences another. The current study explores the application of normalized transfer entropy to EEG data to detect and identify the patterns of information flow in the functional brain networks during cognitive activity. Using a mix of signal processing, information and graph-theoretic techniques, this study has identified and characterized the changing connectivity patterns of the directed functional brain networks during different cognitive tasks. The results demonstrate not only the value of transfer entropy in evaluating the directed functional brain networks but more importantly in determining the information flow patterns and thus providing more insights into the dynamics of the neuronal clusters underpinning cognitive function.

Change Detection and Visualization of Functional Brain Networks using EEG Data

Abstract Mining dynamic and non-trivial patterns of interactions of functional brain networks has gained significance due to the recent advances in the field of computational neuroscience. Sophisticated data search capabilities, advanced signal processing techniques, statistical methods, complex network and graph mining algorithms to unfold and discover hidden patterns in the functional brain network supported with efficient visualization techniques are essential for making potential inferences of the results obtained. Visualization of change in activity during cognitive function is useful to discover and get insights into the hidden, novel and complex neuronal patterns and trends during the normal and cognitive load conditions from the graph/temporal representation of the functional brain network. This paper explores novel methods to detect and track the dynamics and complexity of the brain function. It also uses a new tool called Functional Brain Network Analysis and Visualization (FBNAV) tool to visualize the outcomes of various computational analyses to enable us to identify and study the changing neuronal patterns during various states of the brain activity using augmented/customisedTopoplots andHeadplots. The change detection algorithm tracks and visualizes the cognitive load induced changes across the scalp regions.These techniques may also be helpful to locate and identify patterns in certain abnormal mental states resulting due to some mental disorders such as stress.

Directed Connectivity Analysis of Functional Brain Networks during Cognitive Activity Using Transfer Entropy

Most previous studies of functional brain networks have been conducted on undirected networks despite the fact that direction of information flow is able to provide additional information on how one brain region influences another. The current study explores the application of normalized transfer entropy (NTE) to detect and identify the patterns of information flow in the functional brain networks derived from EEG data during cognitive activity. Using a combination of signal processing, information and graph-theoretic techniques, this study has identified and characterized the changing connectivity patterns of the directed functional brain networks during different cognitive tasks. The functional brain networks constructed from EEG data using non-linear measure NTE also exhibit small-world property. An exponential truncated power-law fits the in-degree and out-degree distribution of directed functional brain networks. The empirical results demonstrate not only the application of transfer entropy in evaluating the directed functional brain networks, but also in determining the information flow patterns and thus provide more insights into the dynamics of the neuronal clusters underpinning cognitive function.

A heuristic branch-and-bound based thresholding algorithm for unveiling cognitive activity from EEG data

Abstract One of the biggest challenges in the field of computational neuroscience from the perspective of complex network analysis is the measurement of dynamic local and global interactions of the brain regions during cognitive function. Graph theoretic analysis has been extensively applied to study the dynamics of functional brain networks in the recent years. The selection of appropriate thresholding methods to construct weighted/unweighted subnetworks to detect cognitive load induced changes in brain׳s electrical activity remains an open challenge in the functional brain network research. This paper reviews the application of statistical and information theoretic metrics to construct the functional brain networks, proposes a novel Branch-and-Bound based thresholding algorithm that extracts the influential subnetwork, and applies efficient computational techniques and complex network metrics to detect and quantify the cognitive activities. The empirical analyses showcase the efficiency of the proposed thresholding algorithm by highlighting the changing neuronal patterns during cognitive activity when compared to that of baseline activity. Statistical evaluation of the results further validates the efficiency of the proposed method as well. The results demonstrate the ability of the proposed algorithm in detecting subtle cognitive load induced changes in functional brain networks.

TVAR modeling of EEG to detect audio distraction during simulated driving

OBJECTIVE The objective of our current study was to look for the EEG correlates that can reveal the engaged state of the brain while undertaking cognitive tasks. Specifically, we aimed to identify EEG features that could detect audio distraction during simulated driving. APPROACH Time varying autoregressive (TVAR) analysis using Kalman smoother was carried out on short time epochs of EEG data collected from participants as they undertook two simulated driving tasks. TVAR coefficients were then used to construct all pole model enabling the identification of EEG features that could differentiate normal driving from audio distracted driving. MAIN RESULTS Pole analysis of the TVAR model led to the visualization of event related synchronization/desynchronization (ERS/ERD) patterns in the form of pole displacements in pole plots of the temporal EEG channels in the z plane enabling the differentiation of the two driving conditions. ERS in the EEG data has been demonstrated during audio distraction as an associated phenomenon. SIGNIFICANCE Visualizing the ERD/ERS phenomenon in terms of pole displacement is a novel approach. Although ERS/ERD has previously been demonstrated as reliable when applied to motor related tasks, it is believed to be the first time that it has been applied to investigate human cognitive phenomena such as attention and distraction. Results confirmed that distracted/non-distracted driving states can be identified using this approach supporting its applicability to cognition research.

A novel pattern mining approach for identifying cognitive activity in EEG based functional brain networks.

The complex nature of neuronal interactions of the human brain has posed many challenges to the research community. To explore the underlying mechanisms of neuronal activity of cohesive brain regions during different cognitive activities, many innovative mathematical and computational models are required. This paper presents a novel Common Functional Pattern Mining approach to demonstrate the similar patterns of interactions due to common behavior of certain brain regions. The electrode sites of EEG-based functional brain network are modeled as a set of transactions and node-based complex network measures as itemsets. These itemsets are transformed into a graph data structure called Functional Pattern Graph. By mining this Functional Pattern Graph, the common functional patterns due to specific brain functioning can be identified. The empirical analyses show the efficiency of the proposed approach in identifying the extent to which the electrode sites (transactions) are similar during various cognitive load states.

Computational Neuroengineering Approaches to Characterise Cognitive Activity in EEG Data

The human brain is one of the most complex and adaptive systems available to society. The brain consists of tens of billion of neurons (processing nodes) and over 100 trillion interconnections. This makes it an extremely complex communication network. The brain functions at a neuronal level have been explored and understood. However, at a systems level, the brain functions relating to “self awareness, conscience, emotion, intelligence, and judgment” still puzzles scientists today. Identifying the integrative aspects of brain structure and function, specifically how the connections and interactions among neuronal elements (neurons, synapses, cerebellum and contextual regions) result in cognition and behavior, is one of the last great frontiers for scientific research. Unraveling the activity of the brain’s billions of neurons and how they combine to form functional networks has been constrained to behavioural observations. It remains further restricted by both technological and ethical constraints; thus, researchers are increasingly turning to sophisticated data search techniques to unravel hidden complexity. Techniques including complex network clustering and graph mining algorithms can be used to further delve into the hidden workings of the human mind. Combining these techniques with advanced signal processing techniques, inferential statistics can be used to support efficient visualization techniques to help researchers unfold and discover hidden patterns and functionality of brain networks. The objective of this chapter is to present an overview of the applications of approaches to multichannel Electroencephalography( EEG) data, bringing together a variety of techniques, including complex network analysis, linear and non-linear statistical methods. These measures include coherence, mutual information, approximate entropy, information visualization, signal processing, multivariate techniques such as the one-way ANalysis Of VAriance (ANOVA), and Post-hoc analysis procedures. The Cognitive Analysis Framework (CAF) approach outlined in this chapter aims to investigate and demonstrate the integration of these techniques and methodologies. The experiments provide deeper understanding of complex brain dynamics as well as allowing the identification of differences in system complexity, believed to underscore normal human cognition.