Song Lai PhD, Presenter: Nothing to Disclose

Shunshan Li, Abstract Co-Author: Nothing to Disclose

John Lackey MS, Abstract Co-Author: Nothing to Disclose

Jianrong Shi MS, Abstract Co-Author: Nothing to Disclose

Jiangxi Xiao, Abstract Co-Author: Nothing to Disclose

Joseph I Tracy PhD, Abstract Co-Author: Nothing to Disclose

Adam Eugene Flanders MD, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

To verify the hypothesis that hemodynamic response onset (HRO), arising from its dependence on cerebral vasculature, is variable in different cortices and subjects.

fMRI experiments were carried out on a 1.5T MRI system. Five male normal subjects (34.0+/-8.2 years of age) were enrolled. Activation paradigm was designed to measure HRO in motor cortex, auditory cortex, and visual cortex in the same fMRI scan, involving 20 s baseline followed by 5 cycles of 20s white noise stimulation alternated with 20s simultaneous bilateral finger tapping and checkerboard pattern visual stimulation (8Hz). In all subjects, slices (6mm) were positioned parallel to AC-PC line. To measure HRO with sufficient temporal resolution, a 4-slice scan was done (TR = 0.38s), containing motor cortex (1 slice), auditory cortex (2 slices) and visual cortex (1 slice) respectively. Whole brain fMRI scans (TR = 2s) were also collected in the same imaging session in each subject using the same activation paradigm. For repeatitability and reproducibility study, each fMRI scan was repeated once in the same imaging session, and each subject was scanned in two imaging sessions that were 1-3 weeks apart. A previously published mutual information-based fMRI signal detection method was applied for fMRI signal analysis.

HRO was found to be significantly different in different subjects and cortices. Using the subject- and cortex-specific HRO measured in the 4-slice scan, significantly more activation pixels were detected in the whole brain fMRI scans.

This study confirmed that fMRI hemodynamic response onset time is subject- and cortex-dependent. It suggests that for optimized fMRI signal detection, it is desirable to incorporate a short TR (<0.5s) scan to measure HRO in the cortices of interest so that the cortex- and subject-specific HRO can be applied to improve activation signal detection in other scans with larger brain coverage.

To verify the hypothesis that hemodynamic response onset (HRO), arising from its dependence on cerebral vasculature, is variable in different cortices and subjects.

fMRI experiments were carried out on a 1.5T MRI system. Five male normal subjects (34.0+/-8.2 years of age) were enrolled. Activation paradigm was designed to measure HRO in motor cortex, auditory cortex, and visual cortex in the same fMRI scan, involving 20 s baseline followed by 5 cycles of 20s white noise stimulation alternated with 20s simultaneous bilateral finger tapping and checkerboard pattern visual stimulation (8Hz). In all subjects, slices (6mm) were positioned parallel to AC-PC line. To measure HRO with sufficient temporal resolution, a 4-slice scan was done (TR = 0.38s), containing motor cortex (1 slice), auditory cortex (2 slices) and visual cortex (1 slice) respectively. Whole brain fMRI scans (TR = 2s) were also collected in the same imaging session in each subject using the same activation paradigm. For repeatitability and reproducibility study, each fMRI scan was repeated once in the same imaging session, and each subject was scanned in two imaging sessions that were 1-3 weeks apart. A previously published mutual information-based fMRI signal detection method was applied for fMRI signal analysis.

HRO was found to be significantly different in different subjects and cortices. Using the subject- and cortex-specific HRO measured in the 4-slice scan, significantly more activation pixels were detected in the whole brain fMRI scans.

This study confirmed that fMRI hemodynamic response onset time is subject- and cortex-dependent. It suggests that for optimized fMRI signal detection, it is desirable to incorporate a short TR (<0.5s) scan to measure HRO in the cortices of interest so that the cortex- and subject-specific HRO can be applied to improve activation signal detection in other scans with larger brain coverage.

Lai, S, Li, S, Lackey, J, Shi, J, Xiao, J, Tracy, J, Flanders, A, et al, , fMRI Hemodynamic Response Onset Time Is Cortex- and Subject-dependent: Implication to Optimization of fMRI Activation Signal Detection and Paradigm Design.  Radiological Society of North America 2005 Scientific Assembly and Annual Meeting, November 27 - December 2, 2005 ,Chicago IL. http://archive.rsna.org/2005/4420483.html