In a first project, we have continued investigating spontaneous brain activity across a broad spectrum of arousal states occurring during natural sleep. Last year, we completed a pilot study in which we collected fMRI, EEG and other physiological measures with the goal of studying differences in brain activity patterns between various sleep depths. We have developed a pre-processing pipeline and removed motion, hear-beat, and respiration related confounds from the signals, and performed a power analysis to prepare for protocol amendment to perform further experiments on an adequate number of additional subjects. Based on this, amended version of protocol has been completed and ready to submit to PIRC/IRB. Data acquired during the sleep study as well as other data from separate experiments and from publicly available fMRI data bases was studied to further examine the relationship between fMRI and systemic physiology, which we previously found to be strongly variable during alertness. The fMRI dependence systemic physiology is a major limitation for the interpretation of fMRI in terms of brain function. This resulted in various important findings. First, we established a strong relationship between peripheral vascular tone and fMRI signal across most brain regions. This result was published in Neuroimage (Ozbay Neuroimage 2018), and interpreted as a potential contribution of previously reported sympathetic vasoconstrictor activity effectuated by an innervation of the cerebral arteries. We then showed that this activity depends strongly on alertness, consistent with the relationship between brain stem arousal and sympathetic activity (Chang ISMRM 2018). While the control of cerebral blood flow by the extrinsic innervation of cerebral arteries is controversial, its confirmation will have substantial impact on the interpretation of fMRI results and the design of future fMRI experiments. We have therefore started to perform further analysis to obtain a better mechanistic understanding of this phenomenon, and look for evidence of a relationship between subcortical arousal, systemic physiology, and fMRI signal. This involved identification of EEG K-complexes, and examine their association with fMRI and pulse-oximeter suignals. In a second project, we have started to develop robust 3D acquisition techniques to allow our novel magnetization transfer (MT) measurement methods to be applied to study myelin loss in MS patients in Dr. Reichs group. 3D methods are desirable because overcome the very limited brain coverage of the 2D acquisition typical of these novel MT approaches. Robust 3D MRI is a challenge, as 3D acquisition is highly sensitive to motion. Preliminary results show that correction of motion-induced variations in magnetic field and RF receive sensitivity is indeed feasible for 3D acquisition. Lastly, in collaboration with the Reich group, we finished a project estimating the relationship between blood volume and blood oxygenation changes during fMRI experiments. This project made use of the availability of volunteers injected with ferumoxitol, a strong intravascular MRI contrast agent. Comparing experiments with and without ferumoxitol injection, we found that blood volume changes preceded blood oxygenation in response to visual stimuli, consistent with previous work. However, the difference was small (< 1s), and generally does not justify administering these (invasive) injections (de Zwart, Neuroimage 2018).
Showing the most recent 10 out of 68 publications
