Published on Tue Jan 28 2020

WISDoM: characterizing neurological timeseries with the Wishart distribution

Carlo Mengucci, Daniel Remondini, Gastone Castellani, Enrico Giampieri

Wishart Distributed Matrices (WISDoM) is a new framework for the quantification of deviation of symmetric positive-definite matrices. The method can be applied to tasks of supervised learning, like classification.

0
0
0
Abstract

WISDoM (Wishart Distributed Matrices) is a new framework for the quantification of deviation of symmetric positive-definite matrices associated to experimental samples, like covariance or correlation matrices, from expected ones governed by the Wishart distribution WISDoM can be applied to tasks of supervised learning, like classification, in particular when such matrices are generated by data of different dimensionality (e.g. time series with same number of variables but different time sampling). We show the application of the method in two different scenarios. The first is the ranking of features associated to electro encephalogram (EEG) data with a time series design, providing a theoretically sound approach for this type of studies. The second is the classification of autistic subjects of the ABIDE study, using brain connectivity measurements.

Wed Apr 10 2019
Computer Vision
Analyzing Dynamical Brain Functional Connectivity As Trajectories on Space of Covariance Matrices
Human brain functional connectivity (FC) is often measured as the similarity of functional MRI responses across brain regions when a brain is either resting or performing a task. This paper aims to statistically analyze the dynamic nature of FC by representing the collective time-series data, over a set of brain regions.
0
0
0
Tue Dec 16 2014
Machine Learning
Estimation of Large Covariance and Precision Matrices from Temporally Dependent Observations
We study the brain functional connectivity using resting-state fMRI data with long-range temporal dependence. We show that several commonly used methods for independent observations can be applied to the temporally dependent data.
0
0
0
Thu Jan 26 2017
Machine Learning
Riemannian-geometry-based modeling and clustering of network-wide non-stationary time series: The brain-network case
This paper advocates Riemannian multi-manifold modeling in the context of network-wide non-stationary time-series analysis. Time-series data, collected progressively over time and across a network, yield features which are viewed as points in or close to a union of multiple submanifolds.
0
0
0
Mon Dec 03 2018
Machine Learning
Large Spectral Density Matrix Estimation by Thresholding
Spectral density matrix estimation of multivariate time series is a classical problem in time series and signal processing. We develop a non-asymptotic theory for regularized estimation of high-dimensional spectral density matrices of Gaussian and linear processes.
0
0
0
Fri Apr 26 2019
Computer Vision
Discovering Common Change-Point Patterns in Functional Connectivity Across Subjects
This paper studies change-points in human brain functional connectivity (FC) FC relates to the similarity of fMRI responses across different brain regions when the brain is simply resting or performing a task. The paper develops a formal statistical test for finding FC change- points in times series associatedwith FC.
0
0
0
Thu Feb 02 2017
Neural Networks
Scaling Properties of Human Brain Functional Networks
We investigate scaling properties of human brain functional networks in the resting-state. Initial studies, based on least-squares fitting, were shown to be inadequate for precise estimation of power-law.
0
0
0