Lesion-Induced Uncoupling of BOLD fMRI Signal from Neuronal Activation

John L. Ulmer, M.D.

Medical College of Wisconsin

Funded in February, 2002: $100000 for 3 years


back to top
John L. Ulmer, M.D.

Associate Professor of Radiology
Medical College of Wisconsin


back to top

The physiological basis of fMRI BOLD signal is the regional vasoactive response induced by neuronal activity. We hypothesize that common cerebral pathologies known to alter normal regional cerebral vasoactivity and cerebral blood flow can uncouple the BOLD signal response from cortical neuronal activity. We hypothesize that this effect can introduce inaccuracies into fMRI generated cortical activation maps.

To develop a model to help understand and correct inaccuracies in fMRI brain mapping data caused by pathologic conditions that are known to alter cerebrovascular physiology.
1. To compare fMRI generated functional visual field maps (FFMap) with behavioral visual field testing, as a model for proving and characterizing the underlying BOLD-neuronal uncoupling, associated with common cerebral conditions including brain tumors, strokes, and AVMs.
2. To show the impact of altered rCBF on BOLD fMRI caused by common cerebral pathologies.
3. To show the impact of altered regional cerebral vasoactivity on BOLD fMRI caused by common cerebral pathologies.

We have chosen to study several common regional cerebral pathologies (brain tumors, AVMs, strokes). FFMapping is a technique that displays the precise spatial correlation of a locus in the visual field to a sub-region of visual cortex responsible for processing that visual information. By directly comparing the FFMap to a behavioral visual field map, we will identify sub-regions of visual cortex where genuine neuronal function is not reflected in a normal BOLD response. FFMap-behavioral comparisons will be made at the sites of regional pathology and in normal contralateral brain. Sub-regions showing a FFMap-behavioral response mismatch with altered signal amplitude, temporal phase, or both compared to the normal corresponding contralateral visual cortex will indicate the presence of pathologically induced BOLD-neuronal uncoupling.


back to top

Ulmer J.L., Hacein-Bey L., Mathews V.P., Mueller W., DeYoe E.A., Prost R., Meyer G., Wascher T.M., Krouwer H.G., Schmainda K.D., and Lowe M.  Lesion-induced pseudo-dominance at functional magnetic resonance imaging: implications for preoperative assessments.  Neurosurgery. 2004 Sep;55(3):569-79.

Salvan C.V., Ulmer J.L., DeYoe E.A., Wascher T., Mathews V.P., Lewis J.W., and Prost R.  Visual object agnosia and pure word alexia: correlation of functional magnetic resonance imaging and lesion localization.  J Comput Assist Tomogr. 2004 Jan-Feb;28(1):63-7.

Schmainda K.M., Rand S.D., Joseph A.M., Ward B.D., Pathak A.P., Baddrudoja M.A., Ulmer J.L., and Krouwer H.G.  Characterization of a first-pass gradient-echo spin-echo method to predict brain tumor grade and angiogenesis.  AJNR Am J Neuroradiol. 2004 Oct;25(9):1524-32.