About this Research Topic
Technological and theoretical advances in neuroscience have provided us with numerous very-large scale data-sets containing details over a wide range of scales, ranging from the atomic, molecular to anatomical, functional and phenotypes. Approaching brain function at this data-intensive multi-layer level is a challenging task: in addition to the intrinsic variability of any disorder, different brains compensate differently for their own intrinsic abnormalities. This might explain why diagnoses -and treatments- for most brain disorders still remain at the behavioral level. In this Research Topic, we seek novel methods in assessing normal and abnormal brain function as well as informing therapeutic interventions.
To tackle such problems we confront two challenges: dimensionality reduction and the problem of data mining, i.e. extracting important features that pertain to the spatio-temporal dynamics of brain activity and its functional connectivity. A fundamental question that then arises is how one can reconstruct low dimensional functional connectivity networks (FCN) from high dimensional data in time and space such as a 4D fMRI experiment, that can reflect in an efficient manner the most-important states and dynamics of brain activity.
State-of-the-art machine and manifold learning techniques have the potential to deepen our understanding into the mechanism that pertain to cognition and its deficits and to lead to breakthroughs in the field of neuroscience regarding early diagnosis, assessment and design of efficient therapies for disorders such as schizophrenia, epilepsy and Parkinson. Tools include, but are not limited to, linear and nonlinear time series analysis of EEG, MEG, fMRI and PET, state-of-the-art numerical analysis for large-scale systems, multi-scale modeling and analysis, connectivity analysis (Granger-based, phase synchrony, mutual information, transfer entropy, etc), and importantly, manifold and machine learning methodologies for handling data complexity (PCA, ICA, ISOMAP, Diffusion Maps, etc).
Specific areas of interest include but are not limited to: the reconstruction of functional networks that govern the spatio-temporal organization of the brain activity, the identification of distinct spatio-temporal patterns that can suggest the mechanisms of specific higher cognitive functions and their deficits, and the modeling and analysis of both task-dependent and resting-state activity from which we still have a lot to learn and discover.
This Research Topic welcomes submissions of both original and reviewing research of the state-of-the-art in the field.
To tackle such problems we confront two challenges: dimensionality reduction and the problem of data mining, i.e. extracting important features that pertain to the spatio-temporal dynamics of brain activity and its functional connectivity. A fundamental question that then arises is how one can reconstruct low dimensional functional connectivity networks (FCN) from high dimensional data in time and space such as a 4D fMRI experiment, that can reflect in an efficient manner the most-important states and dynamics of brain activity.
State-of-the-art machine and manifold learning techniques have the potential to deepen our understanding into the mechanism that pertain to cognition and its deficits and to lead to breakthroughs in the field of neuroscience regarding early diagnosis, assessment and design of efficient therapies for disorders such as schizophrenia, epilepsy and Parkinson. Tools include, but are not limited to, linear and nonlinear time series analysis of EEG, MEG, fMRI and PET, state-of-the-art numerical analysis for large-scale systems, multi-scale modeling and analysis, connectivity analysis (Granger-based, phase synchrony, mutual information, transfer entropy, etc), and importantly, manifold and machine learning methodologies for handling data complexity (PCA, ICA, ISOMAP, Diffusion Maps, etc).
Specific areas of interest include but are not limited to: the reconstruction of functional networks that govern the spatio-temporal organization of the brain activity, the identification of distinct spatio-temporal patterns that can suggest the mechanisms of specific higher cognitive functions and their deficits, and the modeling and analysis of both task-dependent and resting-state activity from which we still have a lot to learn and discover.
This Research Topic welcomes submissions of both original and reviewing research of the state-of-the-art in the field.
Keywords: data-driven analysis, machine learning, manifold learning, brain functional connectivity, pathophysiology
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
