fMRIと拡散強調画像を用いた脳の構造と機能の同時分析

The overall goal of the following experiment is to simultaneously analyze brain structure and function using magnetic resonance imaging. This is achieved using high field MRI to image the brain's white matter structure with diffusion spectrum imaging or DSI and to measure brain function with bold FMRI. The DSI data are then processed to produce multi-directional diffusion estimates at every point in the brain.

Additionally, the FMRI data are analyzed in order to produce regions of interest for generating or selecting virtual white matter fibers. Next, the regions of interest are aligned with the DSI data, so that functional and structural data are in a common image space. Finally, t tractography is performed on diffusion data in order to estimate white matter pathways connecting functional regions of interest Results are obtained that show the degree of anatomical connectivity between brain areas hypothesized to be functionally connected.

Based on FMRI task data. Recent converging evidence has suggested that complex cognitive operations are executed by networks of many brain regions working in concert rather than a single unitary area. In order to fully specify these computational systems, it is necessary to understand the relationship between their functional and structural properties by combining functional MRI.

With diffusion weighted MR Imaging, one can examine network connectivity and how this gives rise to complicated human behavior. The main advantage of this diffusion MRI pipeline over standard methods like diffusion tensor imaging is that the combination of high angular resolution diffusion weighted imaging and model free reconstruction enables us to better resolve complex fiber configurations in the brain. The implications of this technique extend towards characterization of neuropsychological conditions.

For example, congenital prosopagnosia in which individuals show impairment in face recognition. Using standard diffusion MRI, it has been shown that white matter fiber tracts proximal to face processing areas are degraded compared to normal controls. By combining structural and functional MRI one can identify structural deficiencies in fiber, specifically connecting nodes in the face processing network.

This method can also be applied in clinical context such as neurosurgical planning. Surgeons use functional mapping to identify gray matter tissue associated with important cognitive functions in order to minimize any incidental damage during surgery. With additional structural information such as diffusion weighted imaging, they can also minimize the damage to critical white matter structures that connect these functional areas.

Generally, individual new to this method will struggle with the fact that no single software package exists to perform all necessary steps of the procedure. Therefore, users must move between multiple programs while maintaining a common image format along with consistent orientation and alignment. Our protocol includes detailed instruction to guide users through this procedure.

Analyses of this type that consider structure and function in combination are natural extinction of functional imaging experiments that have identified Coved brain areas. In tasks of interest, most previous approaches have not been able to provide information about structural connectivity, and that is what we add in our approach here In this protocol. A Siemens three Tesla scanner is used to acquire a 257 direction diffusion spectrum imaging or DSI scan with a 32 channel phased array head coil, the high field strength and 32 channel coil are required to achieve the signal for this high angular resolution scan.

The most frequently used diffusion weighted imaging method is diffusion tensor imaging or DTI using a five to 10 minute scan typically measuring 64 or fewer directions. A limitation of DTI is its difficulty in resolving, crossing and kissing fibers, which are better detected with a combination of high resolution acquisition and reconstruction methods such as DS.I note that the DSI protocol requires approximately 45 to 50 minutes of imaging, and that motion correction cannot be applied to DSI data. Therefore, it is advisable to minimize movement through the use of bite bars, foam padding, or other stabilization techniques, and to use highly trained participants, additional equipment is required for task-based FMRI, such as an MR compatible display and button response system.

Before scanning, be sure to obtain informed consent and screen for MR.Contraindications. Then brief the participant on the nature of the scans to be performed, emphasizing the need to remain still during the DSI scan. Once the participant is ready to begin comfortably stabilize the participant's head and then slide the bed into the scanner, perform initial scout scans and calibration.

Then align the slices for the DSI scan to the anterior and posterior commissures and ensure that the slices for the DSI scan cover the whole brain run. The DSI scan while the subject relaxes in the scanner or watches a movie on the presentation system. After the DSI scan, collect a T one weighted anatomical scan for later use in coregistering the DSI data with other anatomical or functional data in the same or separate scan session.

Also acquire task-based FMRI data for functional scanning of behavioral tasks. Instruct subjects to monitor the screen for task relevant stimuli and to as required. If FMRI is performed on a separate day, obtain another T one weighted anatomical scan.

This processing approach utilizes surface based analysis of FMRI data to generate ROIs for tractography and allows for better visualization of correspondences between tractography endpoints and functional ROIs. To begin processing, first, submit the acquired T one weighted image to free surfers automated algorithm, which performs anatomical segmentation of gray and white matter and cortico surface reconstruction. Output also includes a processed version of the anatomical volume from which the surfaces were created, referred to as the surface volume.

Next pre-process FMRI data in a acne. Then import the free surfer output into summa a acne software and map pre-processed functional data onto resulting surfaces. Analyze FMRI data to generate statistical maps from which functionally defined ROIs for T tractography can be created.

Then expand these surface based functional ROIs into white matter by dilation to maximize contact with streamlines during tractography. Finally, transform the dilated ROIs from surface to volume coordinates and output as nifty files to process the diffusion data. First, identify which DICOM images in the dataset other B zero or baseline images, and convert these to nifty format.

Next in DSI studio, open the DSI DICOM images and combine to create a source file and supply a gradient table. Next, apply the default reconstruction mask to the baseline image and ensure that it encompasses all gray matter without including empty space, skull or non brainin tissue. Edit the mask necessary.

Choose a high resolution reconstruction model using A-D-S-I-G-Q-I or GQI variance here. The GQI option is used. Then create a fiber information file to represent the principle diffusion directions in each vole.

Next functional ROIs must be transformed to DSI space. Use apni to align the DSIB zero image to the nifty format anatomical surface volume. Invert the resulting 12 point ALINE transformation matrix using the a acne program cat mat.

Then apply the inverted matrix to functional ROIs to transform them into DSI space. Tracking fibers with a whole brain seed is a fast and effective way to assess overall data quality. It also presents an opportunity to determine values for global parameters such as the tracking threshold to begin, create a whole brain seed region.

Then set an initial tracking threshold value to mask out low signal voxels, as well as the angle threshold. Also, set the tracking step size in millimeters and the desired number of fibers or seed points. Now perform whole brain tractography to check overall ODF reconstruction quality.

Next, find an optimal tracking threshold by iteratively performing whole brain tracking and adjusting the tracking threshold. Find a threshold that maximizes the proportion of fibers that reach gray matter by visualizing overlap of whole brain tractography and a gray matter mask in track, fizz noisy fibers are minimized when 90 to 100%of fibers reach gray matter Further, check that the tracking threshold masks out voxel and empty space. For example, the longitudinal fissure without removing voxel, which clearly lie in white matter as a crosscheck track, a set of control fibers from an anatomical ROI at the occipital pole with a large number of seeds, for example, 500, 000.

Check that this procedure produces approximately the same number of fibers across data sets now that the optimal tractography parameters have been chosen. Next, perform ROI constrained T tractography to test hypotheses regarding connectivity between functionally defined brain regions. Begin by loading the fib file and create a whole brains seed region in DSI Studio Next load one or more functionally defined region of interest nifty files, and set them as ROIs in DSI studio region setters.

ROIs will require streamlines to pass through them, set the tracking and angle threshold using previously optimized parameters and perform tracking. Finally, save the tractography output as TRK files. Next, perform endpoint density analysis, which can measure structural connectivity correspondences with precise spatial locations of task-based functional activation.

To begin load the nifty ROIs and TRK files into track fz software, perform Boolean operations between regions and save the results of each operation as a new TRK file. Use the diffusion toolkit functions to spatially transform TRK files from DSI space to surface volume space in order to view fiber data over a high resolution anatomical underlay load the transform TRK file and surface volume in track vis to inspect the results as one measure of connectivity. Calculate the total number of fiber endpoints in an ROI normalized by ROI volume.

Here we see an illustration of optimal and suboptimal results using whole brain tractography. All three images are based on the same 257 direction DWI dataset from a single participant. Optimal results are shown here.

In contrast, the results seen here show the effect of excessively lenient t tractography parameters. Here we see the reduction in quality that results from using a single tensor model to reconstruct the DWI data. In this figure, we see an example of regions activated during a face perception task in which pictures of faces and everyday objects were viewed.

While undergoing FMRI scanning two ventral temporal regions in the middle, fusiform gyrus and inferior occipital gyrus showed significantly greater bold responses for faces than for objects. The figure scene here shows the connections between visual cortex, sensory regions, and a region of attentional control in the posterior parietal cortex. This panel shows the approximate locations of V one, V two and V three seed regions in red, green, and blue respectively.

The PPC seed region labeled IPS one and the fiber tracks that connect these regions tracts are colored by the occipital ROI from which they were seated. Panel B shows the functionality defined regions in IPS in brown, V one in red, V two in green, and V three in blue on the cortical surface along with the fiber endpoints in each region. Once mastered data acquisition for a single participant can be achieved in 30 to 90 minutes.

The automated anatomical surface reconstruction typically takes 16 hours while diffusion weighted data can be processed in less than an hour. The time to process and analyze FMRI data varies depending on the behavioral task and the experimental procedures. Tractography time requirements also range from minutes to hours, depending on the tracking parameters and region of interest constraints.

While attempting this procedure, it's important to remember that tractography result may be susceptible to both false positives and false negatives. Always evaluate your fiber tracking result in the context of prior neuro anatomical findings, or use converging methodologies such as functional connectivity analysis Following this procedure. Other methods like pattern classification on fiber locations, detailed spatial analysis of endpoint distributions and longitudinal scanning of white matter integrity can be performed in order to further investigate the relationship with brain structure and function.

This technique has paved the way for researchers in the field of cognitive neuroscience to explore structure function relationships non-invasively in healthy humans and clinical populations. Structural connectivity between brain regions can serve to constrain hypotheses about the flow of information through brain networks that control complex human behaviors. After watching this video, you should understand the key steps in reconstructing diffusion weighted imaging data and performing fiber tractography.

You should also understand the importance of performing quality checks and iterative parameter tests to optimize your fiber tracking results. Finally, after watching this video, you should have a better understanding of how to relate anatomical connectivity to functional properties of brain networks.