Our Research Projects

Blood Flow Effects
Some interesting work is underway investigating the effect of caffeine (a cerebral vasoconstrictor) on the brain, which may have the potential to be an fMRI contrast booster. The Parrish Lab was the first to explore the effects of caffeine on the BOLD signal in humans. The early work demonstrated the dramatic gain (30-50% increase) in BOLD signal obtained by administering a dose of caffeine equivalent to just two to three cups of coffee. Read Mulderink Paper. However, caffeine is a vasoconstrictor which causes the resting perfusion level to decrease by roughly 20 to 30 percent. This led us to investigate some of the metabolic effects knowing that we perform better with caffeine but have reduced blood flow.

Metabolic Effects
Additional research conducted in the lab demonstrated that the brain physiology becomes more efficient at processing oxygen in the presence of caffeine. This figure demonstrates how the brain becomes more efficient in the presence of caffeine. That is the same level of metabolism post-caffeine requires a smaller blood flow change compared to the pre-caffeine condition. 

Effects on Resting State
Current work is aimed at better understanding caffeine’s effect on the central nervous system with more complex tasks and additional information from other imaging modalities. Another topic of interest is how caffeine interacts with resting state connectivity measured with BOLD. This work demonstrates that there is dramatic change in connectivity measured by time course correlations of resting state data in different networks. These figures indicate that the BOLD resting state signal is a mixture of neuronal information complicated by physiologic noise, which is reduced in the caffeine condition. More work is required to better understand the meaning of resting state BOLD.  (Read Caffeine CMRO2 Paper and Caffeine Dose Paper).

Future studies are aimed at investigating how caffeine  can be used as a cognitive enhancer by facilitating learning or priming the cortex to receive therapy.

DTI Basics
DTI exploits a property of MRI that makes the image sensitive to microscopic movement of water. As a result it is possible to explore white matter connectivity and physiologic properties. Beautiful images of localized white matter tracts are possible as well as being quantitative.

We have utilized DTI metrics in the study of stroke subjects pre/post treatment to demonstrate the effect of aggressive treatment in chronic stroke. In addition, we are using DTI metrics to predict which stroke subjects would respond to therapy. 

DTI for Parkinson’s Disease
A current study in collaboration with colleagues in the Department of Neurology is investigating the use of DTI metrics as a biomarker for Parkinson’s Disease.

DTI of Learning
An exciting study that is in the early stages of development is the effect of learning on the cortical spinal tracts. A study with collaborators at the Rehabilitation Institute of Chicago utilize a Brain Machine Interface to investigate the plasticity of the normal brain. Normal subjects were trained to use their shoulder muscles to guide a 3D cursor using an infrared interface to map the body position to a computer cursor movement. We used Tract based spatial statistics (TBSS) to calculate differences in white matter metrics compared to baseline. The results indicate an increase in the white matter coherence (FA value) in only the non-dominant corticospinal tract. This may be in response to a finer tuning of the non-dominant motor system.

These results demonstrate that it is possible to identify regions of the white matter that are involved in learning. Future work will investigate these same tasks in spinal cord injury patients.

The signal detected in functional magnetic resonance imaging (fMRI) is on the order of a one percent (often less) from the baseline signal. Time series data with a high static and temporal signal-to-noise ratio (SNR) is required to reliably detect these changes. This project developed a model to determine the minimum required signal-to-noise ratio necessary for detecting the expected BOLD signal change.

The immediate benefits are clinical, addressing the neurosurgeon’s need to know if a region of brain near a lesion is active. Knowing the temporal signal-to-noise characteristics allows one to make a more informed decision about the derived functional maps, which is of use to neuroscientists as well. The result is a blood oxygen level dependent (BOLD) sensitivity map that depicts the minimum BOLD signal change that can be detected given the experimental parameters and statistical requirements.(Read TSNR Paper)

More and more studies are involving multiple sites to collect larger diverse populations for functional imaging studies.

Specifically aimed at a disease (Alzheimer’s or Parkinson’s) or comparing cultural differences, it is critical to understand the differences from site to site. Use of temporal signal noise measures allows investigators to identify suspect data or sites. Our group has been using this measure in a number of studies to promote uniform data collection or detect hardware or environmental issues affect the scanning data. 

Another useful tool for investigating the underlying vascular physiology is the impulse response function or the hemodynamic response function. Using brief visually cued motor stimuli, it is possible to measure properties of the local cerebrovasculature. This information is useful for normalizing blood oxygen level dependent (BOLD) signal or for detecting abnormalities that might effect interpretation. We have shown that the HRF can indicate when the cerebrovasculature has returned to normal following treatment (Read Carotid Occlusion Paper).

The HRF measure is a response to a task based challenge. It is important to measure in subjects with known pathology such as a stroke or in an aging population since their vessels often have different properties than normal subjects. In a study with collaborators in the School of Speech and Communication Disorders, we demonstrated that using the incorrect HRF in stroke patients would often mask the true activation of the language network (Read Stroke HRF paper). We directly measured the HRF in different regions of the brain which allowed us to detect language activation in brain tissue adjacent to the stroke that was not previously detected.

The HRF can also be used to monitor the effects of therapy on patients. Working with collaborators in Neurosurgery, we demonstrated changes in the HRF follwoing treatment of a patient with a large temporal lobe arteriovenous malformation (AVM, a congenital condition where arteries bypass the capillary bed and drain directly into the veins). In this case, endovascular treatment was used to access the deep vessels of the AVM to minimize the difficulty of the eventual surgery. Mapping the HRF function demonstrated the truly profound effect the treatment had on the patient’s cerebrovasculature.

This relatively new method of non-invasively mapping brain function uses strong magnetic pulses to stimulate underlying cortex. The breakthrough in this technology has been the marriage of infrared neuronavigation with the TMS device (navigated TMS or nTMS). Now it is possible to know precisely where you are stimulating in the brain. 

Nexstim and the Parrish Lab have been collaborating to compare nTMS presurgical mapping to our clinical presurgical fMRI mapping using direct cortical stimulation intraoperatively as the gold standard. To date, the results of the study have shown that both methods match closely to the intraoperative findings. However, TMS is much less difficult to perform and does not require the patient to lie in a magnet. In fact, the case shown to the right demonstrates how we could not get a successful block motor fMRI study on a 10 year old AVM patient. However, the nTMS study was successful because he was able to watch Star Wars the Clone Wars, while being tested. Since TMS measures the tracts passively, the subject does not have to pay attention or even be awake 

Another exciting study completed by high school students from the Illinois Math and Science Academy investigated the effect of caffeine on motor excitability. In their study, the students used navigated TMS to measure the response to a number of different stimulation levels pre and post caffeine. Due to the ability to precisely locate the magnetic stimulation pulse on the Nexstim system, it was possible to confidently stimulate at the same site. The results indicate that there is a greater than 50 percent increase in the excitability of the motor neurons at the motor threshold level following caffeine and even a larger response at higher  stimulation levels (still sub-threshold for overt movement). 

TMS measures purely neuronal stimulation compared to blood flow based effects in fMRI. Using these modalities together they give us a much better understanding of the complete effects of caffeine on the brain.

Future work will focus on mapping language areas for presurgical mapping and the investigation of therapy in aphasia subjects.

Acupuncture is a treatment for pain and illness in which thin needles are positioned just under the surface of the skin at special centers around the body. It originated in China over 3000 years ago, but despite its longevity Western medicine has been reluctant to accept acupuncture as a valid method of treatment. 

In our lab, we investigated the effect of acupuncture on the brain using functional magnetic resonance imaging (fMRI). Test subjects were notified that they would receive some form of acupuncture during the imaging but were unfamiliar with the procedure. The resulting functional maps generated from group data show that when the visual acupuncture point is stimulated there is bilateral auditory cortex activation. A control point was used that is related to gustation. The results from needling this point did not activate either the visual or auditory regions. This work investigated the time course of the activation, the effect of a combination of points and the activation induced by sham points. (Read IEEE Acupuncture Paper).

Normal CVR 
CVR is a quantitative measure of the cerebrovascular response to a challenge. Typically the challenge is breathing CO₂ enriched air or being administered Acetazolamide, which causes an increase in the cerebral blood flow (CBF). We have implemented a method of breath holding to act as the vascular challenge. These short 15 second breath holds are done while collecting high resolution BOLD sensitive data. Averaging six trials of breath holding allows us to generate a map of the CVR in five minutes. We are using this to investigate the alterations of the cerebrovasculature in stroke subjects, Alzheimer’s disease patients, tumor patients, AVM patients, and subjects enrolled in therapy trials.

Abnormal CVR
The presence of the AVM alters the cerebrovasculature and it is possible to use the CVR map to show the impact of a large occipital AVM. Understanding the CVR will allow us to better detect physiologic changes which may indicate response to therapy or treatment. Additionally, the CVR can be used to normalize subject’s BOLD data in longitudinal or multi-center trials, which is important for improving results.

Voxel-based Lesion Symptom Mapping (VLSM) is able to statistically assess the lesion's effect on behavioral scores on a voxel-by-voxel basis. In each voxel a statistical test is conducted to determine if a difference in the behavioral measures exists between the lesioned and non-lesioned group.

The groups are reassigned at each new voxel location based on the presence of the lesion. A t-test produces a statistical map overlaid on the lesioned voxels. This voxel-by-voxel analysis produces high resolution statistical maps (1 mm³), rather than a large region of interest analysis or lesion categories as seen in prior lesion studies.

VLSM analysis is not limited by tissue type as in voxel based morphometry or BOLD functional MRI which only assesses blood flow changes in gray matter. All 3D lesion maps were entered into VLSM analysis. Areas showing significant correlations with functional performance measures were identified using the false discovery rate corrected at p ≤ 0.05.