MyConnectome fMRI QA Overview

Quality assurance metrics were computed as described by http://preprocessed-connectomes-project.github.io/quality-assessment-protocol/index.html using Chris Gorgolewski's enh branch https://github.com/chrisfilo/quality-assessment-protocol/tree/enh (note: descriptions of QA measures below are copied directly from the QAP web site).

Data from my connectome project were compared to data for the CORR project obtained from https://raw.githubusercontent.com/preprocessed-connectomes-project/quality-assessment-protocol/master/poster_data/corr_func.csv

The following preprocesing was performed:

  1. motion correction: mcflirt using sinc interpolation
  2. compute mean functional from mcflirt output
  3. bet on mean functional to obtain functional brain mask

Summary statistics

myc_mean myc_sd corr_mean corr_sd
Entopy Focus Criterion [EFC] 0.8917690 0.0013351 0.5045509 0.0584542
Foreground-Background Energy Ratio [FBER] 87.8371180 8.8681377 94.4751823 35.5025402
Smoothness[FWHM] 3.1258442 0.0329106 2.0264664 0.2933007
Ghost to Signal Ratio [GSR] 0.0119624 0.0045142 0.0300707 0.0164544
Mean Fractional Displacement 0.0473192 0.0049754 0.0098057 0.0119603
Percentage of vols with FD>0.2 0.6044310 0.7442370 3.7161675 3.1925562
Median Distance Index 0.0290078 0.0075215 16.9233880 22.2620085

Make timeseries plots for each variable across sessions with marginal histograms

plot of chunk tsplot plot of chunk tsplot