Skip to content
No description, website, or topics provided.
MATLAB C Other
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
liblinear-2.11
.gitignore
.swp
BrodmannLabels.m
README.md
beta_binomial_cred_interval.m
brewermap.m
compute_features.m
compute_mad.m
compute_mssd.m
confusion2latex.m
finn_et_al_analysis.m
format_data_for_OSF.m~
getLobeSets.m
get_jobs_run.m
get_new_roi_index_into_old_rois.m
get_results.bash
load_dataset_names.m
lowess.m
plot_1NN_heatmaps.m
plot_BV_heatmap.m
plot_BV_rest_vs_task.m
plot_LR_all_accuacies.m
plot_ROI_mask.m
plot_all.m
plot_confusion_matrices.m
plot_feature_selection_acc.m
plot_finn_all_accuracies.m
preparedatastruct2.m
preprocessROIdata_v1.m
preprocessing_variation_analysis.m
print_result_table_regularization_optimized.m
run_all_feature_selection.m
run_classifiers_in_parallel.m
run_feature_selection_models.m
runclassifier_1NN.m
runclassifier_fully_parallelized.m
set_directories.m
tasklist.mat
tight_subplot.m

README.md

Predicting Task and Subject Identity using Functional Connectivity and BOLD Variability

This repository provides code for the analyses used in the manuscript: Predicting Task and Subject Identity using Functional Connectivity and BOLD Variability submitted to Brain Connectivity.

Preprocessing

The main functions for preprocessing the data and computing features are:

preprocessROIdata_v1.m: formats the time course for further analysis and calls preparedatastruct2.m
compute_features.m: computes and saves machine learning features so that we don't have to recompute features every modeling run

Analysis

The main functions for running the analyses are:

runclassifier_fully_parallelized.m: runs the logistic regression analyses
runclassifier_1NN.m: runs the single nearest neighbor analyses

These functions are called by run_classifiers_in_parallel.m

Results

The main functions for result aggregation and plotting are :

finn_et_al_analysis.m: performs the analyses in Finn et al. 2015
finn_plot_all_accuracies.m: produces bar plots comparing accuracies for each preprocessing variation
plot_all.m: generates plots based on the type of analysis ran (task classification, subject classification, data exploration, preprocessing robustness)

You can’t perform that action at this time.