Meeting Banner
Abstract #2846

Consistent Activation Across Trials and Field Strengths by ROC-Reproducibility Thresholding

Tynan Reid Stevens1, David B. Clarke2, Ryan CN D'Arcy1, Steven D. Beyea1

1IBD-Atlantic, NRC, Halifax, Nova Scotia, Canada; 2QEII Health Science Centre, Halifax, Nova Scotia, Canada

Setting activation thresholds remains a significant challenge in fMRI. Fixed threshold levels do not account for individual variability, habituation, or acquisition strategies. We developed a method for determining data-driven thresholds levels based on test-retest ROC-reproducibility (ROC-r) analysis. Eight subjects performed test-retest fMRI at two field strengths (1.5 & 4 Tesla). Optimized ROC-r thresholds were calculated for the template (5.27 +/- 0.22) and retest images (3.71 +/- 0.18). The ROC-r thresholds were significantly higher for 4 Tesla than 1.5 Tesla. Compared to FDR or Bonferroni thresholds, ROC-r thresholds produced more consistent activation extent across field strengths and trials.

Keywords

ability according account activated activation active addition albeit allowed analyses applying approaches asked background best blocks blue buttons challenge cognitive collected combination computed condition conditions congruent consecutive consisted consistent contained control correct corrected cortical created curvature curve curves datasets detecting determined digit drank driven eddy eliminate error example extent face feather field fixed greater green habituation hand hardware health healthy identify illustrated incongruent incorrect individual language like likely linear makes maps math matrix model motion motor negative neural noise nova operations optimal optimized overlaid paced pair pairs participant participants particularly patterns plateau plotted pressing presumably problematic processes produce produced produces producing profile profiles quantitatively registered reliable replications reproducibility resolution respectively respond rest retest rise scanned scanner scanning select sense sensitivity sentence sentences session sessions setting slightly spatially specificity spiral statistic stimuli strength strengths stroked subject subjects suggests tapping task tasks template templates threshold thresholding thresholds trajectory trials twice typically variability whereas