2018/02/28 by Lea Waller, Anastasia Brovkin, Waller, Lea +9
Environmental Science · Neuroscience · Psychology · #Applications (stat.AP) #FOS: Computer and information sciences #Functional Brain Connectivity Studies #Health, Environment, Cognitive Aging #Mental Health Research Topics
paper · pdf · doi:10.48550/arxiv.1803.00082
openalex publication_date 2018/02/28 · openalex created_date 2022/10/03 · openalex updated_date 2026/07/28
Background: We previously presented GraphVar as a user-friendly MATLAB\ntoolbox for comprehensive graph analyses of functional brain connectivity. Here\nwe introduce a comprehensive extension of the toolbox allowing users to\nseamlessly explore easily customizable decoding models across functional\nconnectivity measures as well as additional features.\n New Method: GraphVar 2.0 provides machine learning (ML) model construction,\nvalidation and exploration. Machine learning can be performed across any\ncombination of network measures and additional variables, allowing for a\nflexibility in neuroimaging applications.\n Results: In addition to previously integrated functionalities, such as\nnetwork construction and graph-theoretical analyses of brain connectivity with\na high-speed general linear model (GLM), users can now perform customizable ML\nacross connectivity matrices, network metrics and additionally imported\nvariables. The new extension also provides parametric and nonparametric testing\nof classifier and regressor performance, data export, figure generation and\nhigh quality export.\n Comparison with existing methods: Compared to other existing toolboxes,\nGraphVar 2.0 offers (1) comprehensive customization, (2) an all-in-one user\nfriendly interface, (3) customizable model design and manual hyperparameter\nentry, (4) interactive results exploration and data export, (5) automated\ncueing for modelling multiple outcome variables within the same session, (6) an\neasy to follow introductory review.\n Conclusions: GraphVar 2.0 allows comprehensive, user-friendly exploration of\nencoding (GLM) and decoding (ML) modelling approaches on functional\nconnectivity measures making big data neuroscience readily accessible to a\nbroader audience of neuroimaging investigators.\n