Melvin's digital garden

Cluster failure: Why FMRI inferences for spatial extent have inflated false-positive rates

[2016-08-29 Mon 20:06:52] speaker: Ivan Vanzaj event: Papers We Love

sensational reporting my major news outlets and on hackernews

types of MRI studies: functional, perfusion, diffusion

functional MRI = MRI + stimulation + response

Blood oxygenation level dependent (BOLD) effect bain activity -> local increase in blood flow -> [oxy/deoxy hemoglobin] -> MR signal change (1-5%)

EPI acquisition to acquire voxels, compare activation of voxels and stimulation pattern

Our assumption of spatial noise as gaussian is incorrect, affects about 3000 studies.

Links to this note