Meeting 2: Summary and slides
Meeting 2 took place on the 15th of May 2012
Presentations & Discussion
Simon: Motion artifacts in drug endophenotypes study with170 subjects
Simon showed that the motion effects in the drug endophenotypes study were not significantly eliminated at higher wavelet scales.
Tiago: Lags and motion in fMRI data
Tiago showed that lags (as determined by cross-correlation) in fMRI data scale with distance and that this could potentially be driving the motion-distance effect observed by Power et al. This mechanisitc approach is interesting, but raises the question of what these observed lags in fact mean physically. MI and other measures which take lags into account may help to factor out such effects.
Ameera: Diagnostics Package
Ameera gave a demo of the diagnostics package so far. She has implemented a parcellation scheme which can be used by everyone (to ensure all ROIs are defined the same way before comparisons across datasets).
Joao: Filtering
Joao has suggested a generalized (weighted) filtering method, which he will be implementing. This will weight correlations by the reliability (as determined by the FD).
Outcomes
- Note that Satterswaithe has a long review article on optimal preprocessing under review at NeuroImage. He was contacted by Mika and has agreed to send us a preprint as soon as the reviewers comments have been addressed.
- Ameera is working on developping a 'diagnostics' package which everyone can simply and efficiently apply to their data to quantify the effects of motion.
- Simon has shared the drugs endophenotype data (see 'benchmark datasets' section)