Statistical and Machine Learning Link Selection Methods for Brain Functional Networks: Review and Comparison

Abstract

Network-based representations have introduced a revolution in neuroscience, expanding the understanding of the brain from the activity of individual regions to the interactions between them. This augmented network view comes at the cost of high dimensionality, which hinders both our capacity of deciphering the main mechanisms behind pathologies, and the significance of any statistical and/or machine learning task used in processing this data. A link selection method, allowing to remove irrelevant connections in a given scenario, is an obvious solution that provides improved utilization of these network representations. In this contribution we review a large set of statistical and machine learning link selection methods and evaluate them on real brain functional networks. Results indicate that most methods perform in a qualitatively similar way, with NBS (Network Based Statistics) winning in terms of quantity of retained information, AnovaNet in terms of stability and ExT (Extra Trees) in terms of lower computational cost. While machine learning methods are conceptually more complex than statistical ones, they do not yield a clear advantage. At the same time, the high heterogeneity in the set of links retained by each method suggests that they are offering complementary views to the data. The implications of these results in neuroscience tasks are finally discussed.

Machine learning models reveal the critical role of nighttime systolic blood pressure in predicting functional outcome for acute ischemic stroke after endovascular thrombectomy

“…To reduce model complexity and the risk of overfitting, as well as optimize training speed, we performed least absolute shrinkage and selection operator (LASSO), support vector machine-recursive feature elimination (SVM-RFE), and Boruta to select potential features. SVM-RFE aims to find an optimal subset of features by iteratively removing the least important features based on their weights or rankings obtained from the SVM model, ensuring a global optimal solution ( 18 ). The Boruta algorithm provides an importance score for each feature, which helps to understand the relative importance between features ( 19 ).…”

Synergistic integration of Multi-View Brain Networks and advanced machine learning techniques for auditory disorders diagnostics

“…Machine learning methods can also be used to develop fMRI network inference methods, such as BrainNET, which quantify the contributions of various brain regions [ 36 ]. Deep learning techniques, such as Graph AuTo-Encoding (GATE), have been devised to characterize the population distribution of brain graphs and infer their relationships with human characteristics [ 37 ]. In addition, deep learning methods have been applied to classify brain networks for detecting Alzheimer’s disease (AD) [ 38 ].…”

NDCN-Brain: An Extensible Dynamic Functional Brain Network Model

“…The fMRI-based BFN are constructed such that brain regions or voxels are used as nodes and the correlations of BOLD changes between them are used as connection conditions, where most of them are static brain functional networks [7,8]. Static networks can reflect the connectivity pattern of the brain, but ignore the timing change information of BOLD signals; however, with the development of temporal graphs, dynamic brain functional networks (D-BFN) gradually enter the field of view of related researchers and have received more and more attention.…”