Skip to content

lillianjisoo/norm

Repository files navigation

Rockland Sample Norm Processing Pipeline

This repository has been created by mnarayan, adimaron, and jisoo.

1. Denoising Images

Note: if you already have denoised derivatives, you may skip this step.

2. Extracting ROI Time-series

To extract the ROI time-series using Schaefer atlases of 100, 200, and 300 parcels:

mkdir $roi_output_dir
cd 02_roi_timeseries && python fetch_roi_timeseries.py -i $denoised_output_dir -o $roi_output_dir

3. Successive Normalization

To successively normalize ROI time-series matrices:

mkdir $normalized_output_dir
cd 03_successive_normalization && bash run_normalizer.sh $roi_output_dir $normalized_output_dir

4. ROI Correlation Analysis

To get the correlation among the normalized ROI timeseries:

mkdir $roi_correlation_output_dir
cd 04_roi_correlation && bash run_correlation.sh $normalized_output_dir $roi_correlation_output_dir

5. Network Connectivity Analysis

To get the average connectivity within each network:

mkdir $network_output_dir
cd 05_network_connectivity && bash run_network_corr.py -i $roi_correlation_output_dir -o $network_output_dir

About

Rockland Norm Analysis

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published