Difference between revisions of "Modules:BRAINSDemonWarp"

From Slicer Wiki
Jump to: navigation, search
(Created page with 'Return to Slicer 3.6 Documentation Gallery of New Features __NOTOC__ ===Module Name=== MyModule {| |[[Image:screen…')
 
m (Text replacement - "https?:\/\/(?:www|wiki)\.slicer\.org\/slicerWiki\/index\.php\/([^ ]+) " to "https://www.slicer.org/wiki/$1 ")
 
(14 intermediate revisions by 3 users not shown)
Line 8: Line 8:
  
 
{|
 
{|
|[[Image:screenshotBlankNotOptional.png|thumb|280px|User Interface]]
+
|[[Image:BRAINSDemonWarpForm.png|thumb|280px|User Interface]]
|[[Image:screenshotBlank.png|thumb|280px|Output]]
 
|[[Image:screenshotBlank.png|thumb|280px|Caption]]
 
 
|}
 
|}
  
Line 16: Line 14:
 
===Module Type & Category===
 
===Module Type & Category===
  
Type: Interactive or CLI
+
Type: CLI
 +
 
 +
Category: Registration
 +
 
  
Category: Base or (Filtering, Registration, ''etc.'')
 
  
 
===Authors, Collaborators & Contact===
 
===Authors, Collaborators & Contact===
* Author1: Affiliation & logo, if desired
+
 
* Contributor1: Affiliation & logo, if desired
+
Author: This tool was developed by Hans J. Johnson and Greg Harris.
* Contributor2: Affiliation & logo, if desired
+
 
* Contact: name, email
+
Contributors: The development of this tool was supported by funding from grants NS050568 and NS40068 from the National Institute of Neurological Disorders and Stroke and grants MH31593, MH40856, from the National Institute of Mental Health. 
 +
 
 +
 
  
 
===Module Description===
 
===Module Description===
Overview of what the module does goes here.
 
  
== Usage ==
+
{| style="color:green" border="1"
  
===Use Cases, Examples===
+
|Program title || BRAINSDemonWarp
  
This module is especially appropriate for these use cases:
+
|-
 +
|Program version || 3.0.0
  
* Use Case 1:
+
|-
* Use Case 2:
+
|Program documentation-url || https://www.slicer.org/wiki/Modules:BRAINSDemonWarp |-
 +
|}
  
Examples of the module in use:
 
  
* Example 1
 
* Example 2
 
  
===Tutorials===
+
== Usage ==
 +
The module registers the template (moving) image onto the physical space of the target (fixed) image. The resultant deformation fields and metric values can be written to a file. The program uses the Insight Toolkit (www.ITK.org) for all the computations, and can operate on any of the image file types supported by that library.
  
Links to tutorials explaining how to use this module:
+
===Use Cases, Examples===
 +
The inputs to the BrainDemonsWarp program are the target image, the template image and the optional parameters. These parameters define the arguments for histogram matching and multi resolution registration. The outputs are the deformation field, output image, checkerboard image of the output and the fixed image and the x,y,z displacement vectors. If we specify debug option we can get the outputs at different stages. The filter is templated over the input image type, real image type and the output image types. We implement the algorithm by parsing the input, preprocessing them and registering the processed images.
  
* Tutorial 1
+
# Parsing - The images are initialized by the ValidationInputParser. This function reads in the arguments from the parameter file. It sets the histogram bins, match points, number of levels in the multi resolution registration, shrink factors and number of iterations at each levels. If the orientations of the images are different it sets the orientation of the moving image to that of the fixed image.
** Data Set 1
+
# PreProcessing - In the next step the DemonsPreProcessor preprocesses the images by resampling the template image to target image space. The intensity mismatch problem is solved by histogram matching the images. Histogram matching is done only if the command line option -e is set. ItkHistogramMatchingImageFilter is used to perform this function. Another important step in preprocessing is skull stripping. Skull stripping is done only if the command line option -maskProcessingMode is set to BOBF. We have written an itk filter, named itkBOBFFilter for this purpose. This filter takes in an input image and a whole brain mask and outputs a Brain Only Background Filled(BOBF) image. The non-brain parts in the image are filled with the user specified background value. All computations are performed in the precision of float data.
 +
# Registration - The resulting moving Image and the fixed image are given as inputs to the demons registrator.It uses the MultiResolutionPDEDeformableRegistration filter with NN extrapolation as interpolator and implements the demons deformable algorithm by computing the deformation field which will map a moving image onto a fixed image. It is assumed that the vector elements behave like floating point scalars. Each vector in the deformation field represent the distance between a geometric point in the input space and a point in the output space. The output image is generated by warping the input image with the deformation field using the ItkWarpImageFilter. WarpImageFilter warps an existing image with respect to a given deformation field. Typically the mapped position does not correspond to an integer pixel position in the input image. Interpolation via an image function is used to compute values at non-integer positions. We have used the LinearInterpolateImageFunction for our application. To write the output image we cast the image to the user specified output pixel type.
  
 
===Quick Tour of Features and Use===
 
===Quick Tour of Features and Use===
  
A list panels in the interface, their features, what they mean, and how to use them. For instance:
+
* <span style="color:blue">'''''Inputs''''' </span> Common input parameters.  All must be specified
 +
** <span style="color:green">'''Moving Image Volume'''</span> [<span style="color:orange">--movingVolume</span>][<span style="color:pink">-m</span>]: Required: input moving image
 +
** <span style="color:green">'''Fixed Image Volume'''</span> [<span style="color:orange">--fixedVolume</span>][<span style="color:pink">-f</span>]: Required: input fixed (target) image
 +
** <span style="color:green">'''Input Type'''</span> [<span style="color:orange">--inputPixelType</span>]: Input volumes will be typecast to this format: float|short|ushort|int|uchar
 +
* <span style="color:blue">'''''Outputs''''' </span> Common output parameters.  At least one must be specified.
 +
** <span style="color:green">'''Output Image Volume'''</span> [<span style="color:orange">--outputVolume</span>][<span style="color:pink">-o</span>]: Required: output resampled moving image (will have the same physical space as the fixedVolume).
 +
** <span style="color:green">'''Output Deformation Field Volume'''</span> [<span style="color:orange">--outputDeformationFieldVolume</span>][<span style="color:pink">-O</span>]: Output deformation field vector image (will have the same physical space as the fixedVolume).
 +
** <span style="color:green">'''Output Type'''</span> [<span style="color:orange">--outputPixelType</span>]: outputVolume will be typecast to this format: float|short|ushort|int|uchar
 +
** <span style="color:green">'''Interpolation Mode'''</span> [<span style="color:orange">--interpolationMode</span>]: Type of interpolation to be used when applying transform to moving volume.  Options are Linear, NearestNeighbor, BSpline, or WindowedSinc
 +
* <span style="color:blue">'''''Registration Parameters''''' </span> Input parameters controlling the behavior of the demons registration.
 +
** <span style="color:green">'''Registration Filter Type'''</span> [<span style="color:orange">--registrationFilterType</span>]: Registration Filter Type: Demons|FastSymmetricForces|Diffeomorphic|LogDemons|SymmetricLogDemons
 +
** <span style="color:green">'''Deformation field smoothing sigma'''</span> [<span style="color:orange">--smoothDeformationFieldSigma</span>][<span style="color:pink">-s</span>]: A gaussian smoothing value to be applied to the deformation feild at each iteration.
 +
** <span style="color:green">'''Pyramid Levels'''</span> [<span style="color:orange">--numberOfPyramidLevels</span>][<span style="color:pink">-n</span>]: Number of image pyramid levels to use in the multi-resolution registration.
 +
** <span style="color:green">'''Moving Shrink Factors'''</span> [<span style="color:orange">--minimumFixedPyramid</span>]: The shrink factor for the first level of the fixed image pyramid. (i.e. start at 1/16 scale, then 1/8, then 1/4, then 1/2, and finally full scale)
 +
** <span style="color:green">'''Moving Shrink Factors'''</span> [<span style="color:orange">--minimumMovingPyramid</span>]: The shrink factor for the first level of the moving image pyramid. (i.e. start at 1/16 scale, then 1/8, then 1/4, then 1/2, and finally full scale)
 +
** <span style="color:green">'''the iteration count for each pyramid level'''</span> [<span style="color:orange">--arrayOfPyramidLevelIterations</span>][<span style="color:pink">-i</span>]: The number of iterations for each pyramid level
 +
* <span style="color:blue">''''' Input image pre-processing options ''''' </span>  Operations to perform on the images before starting the registration.
 +
** <span style="color:green">'''Histogram Match'''</span> [<span style="color:orange">--histogramMatch</span>][<span style="color:pink">-e</span>]: Histogram Match the input images.  This is suitable for images of the same modality that may have different absolute scales, but the same overall intensity profile.
 +
** <span style="color:green">'''Number of histogram bins to use in histogramMatch'''</span> [<span style="color:orange">--numberOfHistogramBins</span>]: The number of histogram levels
 +
** <span style="color:green">'''Number of Match Points'''</span> [<span style="color:orange">--numberOfMatchPoints</span>]: The number of match points for histrogramMatch
 +
** <span style="color:green">'''Median Filter Size'''</span> [<span style="color:orange">--medianFilterSize</span>]: Median filter radius in all 3 directions.  When images have a lot of salt and pepper noise, this step can improve the registration.
 +
* <span style="color:blue">''''' Deformation field initialziation ''''' </span>  Options for initializing the deformation field.
 +
** <span style="color:green">'''Initial Deformation Field Volume'''</span> [<span style="color:orange">--initializeWithDeformationField</span>]: Initial deformation field vector image file name
 +
** <span style="color:green">'''Initial Transform  Filename'''</span> [<span style="color:orange">--initializeWithTransform</span>]: Initial Transform filename
 +
* <span style="color:blue">''''' Mask Options ''''' </span>  Options for setting binary masks for the registration.
 +
** <span style="color:green">'''Mask Processing Mode'''</span> [<span style="color:orange">--maskProcessingMode</span>]: What mode to use for using the masks: NOMASK|ROIAUTO|ROI|BOBF.  If ROIAUTO is choosen, then the mask is implicitly defined using a otsu forground and hole filling algorithm. Where the Region Of Interest mode uses the masks to define what parts of the image should be used for computing the deformation field. Brain Only Background Fill uses the masks to pre-process the input images by clipping and filling in the background with a predefined value.
 +
** <span style="color:green">'''Input Fixed Mask Filename'''</span> [<span style="color:orange">--fixedBinaryVolume</span>]: Mask filename for desired region of interest in the Fixed image.
 +
** <span style="color:green">'''Input Moving Mask Filename'''</span> [<span style="color:orange">--movingBinaryVolume</span>]: Mask filename for desired region of interest in the Moving image.
 +
** <span style="color:green">'''BOBF Lower Threshold'''</span> [<span style="color:orange">--lowerThresholdForBOBF</span>]: Lower threshold for performing BOBF
 +
** <span style="color:green">'''BOBF Upper Threshold'''</span> [<span style="color:orange">--upperThresholdForBOBF</span>]: Upper threshold for performing BOBF
 +
** <span style="color:green">'''BOBF Background Replacement Value'''</span> [<span style="color:orange">--backgroundFillValue</span>]: Replacement value to overwrite background when performing BOBF
 +
** <span style="color:green">'''Seed For BOBF'''</span> [<span style="color:orange">--seedForBOBF</span>]: coordinates in all 3 directions for Seed when performing BOBF
 +
** <span style="color:green">'''Seed For BOBF'''</span> [<span style="color:orange">--neighborhoodForBOBF</span>]: neighborhood in all 3 directions to be included when performing BOBF
 +
* <span style="color:blue">''''' Advanced Options ''''' </span>  Uncommon debugging or other advanced options.
 +
** <span style="color:green">'''Component displacement field prefix'''</span> [<span style="color:orange">--outputDisplacementFieldPrefix</span>]: Displacement field filename prefix for writing separate x, y, and z component images
 +
** <span style="color:green">'''Output Checkerboard Volume'''</span> [<span style="color:orange">--outputCheckerboardVolume</span>]: Genete a checkerboard image volume between the fixedVolume and the deformed movingVolume.
 +
** <span style="color:green">'''Checkerboard Pattern Subdivisions'''</span> [<span style="color:orange">--checkerboardPatternSubdivisions</span>]: Number of Checkerboard subdivisions in all 3 directions
 +
** <span style="color:green">'''Warp Normalized Images'''</span> [<span style="color:orange">--outputNormalized</span>]: Flag to warp and write the normalized images to output.  In normalized images the image values are fit-scaled to be between 0 and the maximum storage type value.
 +
** <span style="color:green">'''Write Intermediate Images'''</span> [<span style="color:orange">--outputDebug</span>][<span style="color:pink">-v</span>]: Flag to write debugging images after each step.
 +
** <span style="color:green">'''Gradient Type'''</span> [<span style="color:orange">--gradient_type</span>][<span style="color:pink">-t</span>]: Type of gradient used for computing the demons force (0 is symmetrized, 1 is fixed image, 2 is moving image)
 +
** <span style="color:green">'''Update field smoothing'''</span> [<span style="color:orange">--upFieldSmoothing</span>][<span style="color:pink">-g</span>]: Smoothing sigma for the update field at each iteration
 +
** <span style="color:green">'''Max Step Length'''</span> [<span style="color:orange">--max_step_length</span>][<span style="color:pink">-l</span>]: Maximum length of an update vector (0: no restriction)
 +
** <span style="color:green">'''Turn off Diffeomorphism'''</span> [<span style="color:orange">--use_vanilla_dem</span>][<span style="color:pink">-a</span>]: Run vanilla demons algorithm
  
{|
 
|
 
* '''Input panel:'''
 
** '''First input'''
 
** '''Second input'''
 
* '''Parameters panel:'''
 
** '''First parameter'''
 
** '''Second parameter'''
 
* '''Output panel:'''
 
** '''First output'''
 
** '''Second output'''
 
* '''Viewing panel:'''
 
|[[Image:screenshotBlankNotOptional.png|thumb|280px|User Interface]]
 
|}
 
  
 
== Development ==
 
== Development ==
  
 
===Notes from the Developer(s)===
 
===Notes from the Developer(s)===
 
Algorithms used, library classes depended upon, use cases, etc.
 
  
 
===Dependencies===
 
===Dependencies===
 
Other modules or packages that are required for this module's use.
 
  
 
===Tests===
 
===Tests===
 
+
http://testing.psychiatry.uiowa.edu/CDash
On the [http://www.cdash.org/CDash/index.php?project=Slicer3 Dashboard], these tests verify that the module is working on various platforms:
 
 
 
* MyModuleTest1 [http://viewvc.slicer.org/viewcvs.cgi/trunk MyModuleTest1.cxx]
 
* MyModuleTest2 [http://viewvc.slicer.org/viewcvs.cgi/trunk MyModuleTest2.cxx]
 
 
 
===Known bugs===
 
 
 
Links to known bugs in the Slicer3 bug tracker
 
 
 
* [http://www.na-mic.org/Bug/view.php?id=000 Bug 000: description]
 
 
 
  
 
===Usability issues===
 
===Usability issues===
  
 
Follow this [http://na-mic.org/Mantis/main_page.php link] to the Slicer3 bug tracker. Please select the '''usability issue category''' when browsing or contributing.
 
Follow this [http://na-mic.org/Mantis/main_page.php link] to the Slicer3 bug tracker. Please select the '''usability issue category''' when browsing or contributing.
 
===Source code & documentation===
 
 
Links to the module's source code:
 
 
Source code:
 
*[http://viewvc.slicer.org/viewcvs.cgi/trunk file.cxx ]
 
*[http://viewvc.slicer.org/viewcvs.cgi/trunk file.h ]
 
 
Doxygen documentation:
 
*[http://www.na-mic.org/Slicer/Documentation/Slicer3-doc/html/classes.html class1]
 
  
 
== More Information ==  
 
== More Information ==  
  
 
===Acknowledgment===
 
===Acknowledgment===
Include funding and other support here.
+
The development of this tool was supported by funding from grants NS050568 and NS40068 from the National Institute of Neurological Disorders and Stroke and grants MH31593, MH40856, from the National Institute of Mental Health.
  
 
===References===
 
===References===
Publications related to this module go here. Links to pdfs would be useful.
+
* [http://hdl.handle.net/1926/1517 BRAINSDemonWarp: An Applicaton to Perform Demons Registration], Johnson H., Zhao Y., Insight Journal, 2009.

Latest revision as of 13:19, 27 November 2019

Home < Modules:BRAINSDemonWarp

Return to Slicer 3.6 Documentation

Gallery of New Features


Module Name

MyModule

User Interface

General Information

Module Type & Category

Type: CLI

Category: Registration


Authors, Collaborators & Contact

Author: This tool was developed by Hans J. Johnson and Greg Harris.

Contributors: The development of this tool was supported by funding from grants NS050568 and NS40068 from the National Institute of Neurological Disorders and Stroke and grants MH31593, MH40856, from the National Institute of Mental Health.


Module Description

Program title BRAINSDemonWarp
Program version 3.0.0
Program documentation-url -


Usage

The module registers the template (moving) image onto the physical space of the target (fixed) image. The resultant deformation fields and metric values can be written to a file. The program uses the Insight Toolkit (www.ITK.org) for all the computations, and can operate on any of the image file types supported by that library.

Use Cases, Examples

The inputs to the BrainDemonsWarp program are the target image, the template image and the optional parameters. These parameters define the arguments for histogram matching and multi resolution registration. The outputs are the deformation field, output image, checkerboard image of the output and the fixed image and the x,y,z displacement vectors. If we specify debug option we can get the outputs at different stages. The filter is templated over the input image type, real image type and the output image types. We implement the algorithm by parsing the input, preprocessing them and registering the processed images.

  1. Parsing - The images are initialized by the ValidationInputParser. This function reads in the arguments from the parameter file. It sets the histogram bins, match points, number of levels in the multi resolution registration, shrink factors and number of iterations at each levels. If the orientations of the images are different it sets the orientation of the moving image to that of the fixed image.
  2. PreProcessing - In the next step the DemonsPreProcessor preprocesses the images by resampling the template image to target image space. The intensity mismatch problem is solved by histogram matching the images. Histogram matching is done only if the command line option -e is set. ItkHistogramMatchingImageFilter is used to perform this function. Another important step in preprocessing is skull stripping. Skull stripping is done only if the command line option -maskProcessingMode is set to BOBF. We have written an itk filter, named itkBOBFFilter for this purpose. This filter takes in an input image and a whole brain mask and outputs a Brain Only Background Filled(BOBF) image. The non-brain parts in the image are filled with the user specified background value. All computations are performed in the precision of float data.
  3. Registration - The resulting moving Image and the fixed image are given as inputs to the demons registrator.It uses the MultiResolutionPDEDeformableRegistration filter with NN extrapolation as interpolator and implements the demons deformable algorithm by computing the deformation field which will map a moving image onto a fixed image. It is assumed that the vector elements behave like floating point scalars. Each vector in the deformation field represent the distance between a geometric point in the input space and a point in the output space. The output image is generated by warping the input image with the deformation field using the ItkWarpImageFilter. WarpImageFilter warps an existing image with respect to a given deformation field. Typically the mapped position does not correspond to an integer pixel position in the input image. Interpolation via an image function is used to compute values at non-integer positions. We have used the LinearInterpolateImageFunction for our application. To write the output image we cast the image to the user specified output pixel type.

Quick Tour of Features and Use

  • Inputs Common input parameters. All must be specified
    • Moving Image Volume [--movingVolume][-m]: Required: input moving image
    • Fixed Image Volume [--fixedVolume][-f]: Required: input fixed (target) image
    • Input Type [--inputPixelType]: Input volumes will be typecast to this format: float|short|ushort|int|uchar
  • Outputs Common output parameters. At least one must be specified.
    • Output Image Volume [--outputVolume][-o]: Required: output resampled moving image (will have the same physical space as the fixedVolume).
    • Output Deformation Field Volume [--outputDeformationFieldVolume][-O]: Output deformation field vector image (will have the same physical space as the fixedVolume).
    • Output Type [--outputPixelType]: outputVolume will be typecast to this format: float|short|ushort|int|uchar
    • Interpolation Mode [--interpolationMode]: Type of interpolation to be used when applying transform to moving volume. Options are Linear, NearestNeighbor, BSpline, or WindowedSinc
  • Registration Parameters Input parameters controlling the behavior of the demons registration.
    • Registration Filter Type [--registrationFilterType]: Registration Filter Type: Demons|FastSymmetricForces|Diffeomorphic|LogDemons|SymmetricLogDemons
    • Deformation field smoothing sigma [--smoothDeformationFieldSigma][-s]: A gaussian smoothing value to be applied to the deformation feild at each iteration.
    • Pyramid Levels [--numberOfPyramidLevels][-n]: Number of image pyramid levels to use in the multi-resolution registration.
    • Moving Shrink Factors [--minimumFixedPyramid]: The shrink factor for the first level of the fixed image pyramid. (i.e. start at 1/16 scale, then 1/8, then 1/4, then 1/2, and finally full scale)
    • Moving Shrink Factors [--minimumMovingPyramid]: The shrink factor for the first level of the moving image pyramid. (i.e. start at 1/16 scale, then 1/8, then 1/4, then 1/2, and finally full scale)
    • the iteration count for each pyramid level [--arrayOfPyramidLevelIterations][-i]: The number of iterations for each pyramid level
  • Input image pre-processing options Operations to perform on the images before starting the registration.
    • Histogram Match [--histogramMatch][-e]: Histogram Match the input images. This is suitable for images of the same modality that may have different absolute scales, but the same overall intensity profile.
    • Number of histogram bins to use in histogramMatch [--numberOfHistogramBins]: The number of histogram levels
    • Number of Match Points [--numberOfMatchPoints]: The number of match points for histrogramMatch
    • Median Filter Size [--medianFilterSize]: Median filter radius in all 3 directions. When images have a lot of salt and pepper noise, this step can improve the registration.
  • Deformation field initialziation Options for initializing the deformation field.
    • Initial Deformation Field Volume [--initializeWithDeformationField]: Initial deformation field vector image file name
    • Initial Transform Filename [--initializeWithTransform]: Initial Transform filename
  • Mask Options Options for setting binary masks for the registration.
    • Mask Processing Mode [--maskProcessingMode]: What mode to use for using the masks: NOMASK|ROIAUTO|ROI|BOBF. If ROIAUTO is choosen, then the mask is implicitly defined using a otsu forground and hole filling algorithm. Where the Region Of Interest mode uses the masks to define what parts of the image should be used for computing the deformation field. Brain Only Background Fill uses the masks to pre-process the input images by clipping and filling in the background with a predefined value.
    • Input Fixed Mask Filename [--fixedBinaryVolume]: Mask filename for desired region of interest in the Fixed image.
    • Input Moving Mask Filename [--movingBinaryVolume]: Mask filename for desired region of interest in the Moving image.
    • BOBF Lower Threshold [--lowerThresholdForBOBF]: Lower threshold for performing BOBF
    • BOBF Upper Threshold [--upperThresholdForBOBF]: Upper threshold for performing BOBF
    • BOBF Background Replacement Value [--backgroundFillValue]: Replacement value to overwrite background when performing BOBF
    • Seed For BOBF [--seedForBOBF]: coordinates in all 3 directions for Seed when performing BOBF
    • Seed For BOBF [--neighborhoodForBOBF]: neighborhood in all 3 directions to be included when performing BOBF
  • Advanced Options Uncommon debugging or other advanced options.
    • Component displacement field prefix [--outputDisplacementFieldPrefix]: Displacement field filename prefix for writing separate x, y, and z component images
    • Output Checkerboard Volume [--outputCheckerboardVolume]: Genete a checkerboard image volume between the fixedVolume and the deformed movingVolume.
    • Checkerboard Pattern Subdivisions [--checkerboardPatternSubdivisions]: Number of Checkerboard subdivisions in all 3 directions
    • Warp Normalized Images [--outputNormalized]: Flag to warp and write the normalized images to output. In normalized images the image values are fit-scaled to be between 0 and the maximum storage type value.
    • Write Intermediate Images [--outputDebug][-v]: Flag to write debugging images after each step.
    • Gradient Type [--gradient_type][-t]: Type of gradient used for computing the demons force (0 is symmetrized, 1 is fixed image, 2 is moving image)
    • Update field smoothing [--upFieldSmoothing][-g]: Smoothing sigma for the update field at each iteration
    • Max Step Length [--max_step_length][-l]: Maximum length of an update vector (0: no restriction)
    • Turn off Diffeomorphism [--use_vanilla_dem][-a]: Run vanilla demons algorithm


Development

Notes from the Developer(s)

Dependencies

Tests

http://testing.psychiatry.uiowa.edu/CDash

Usability issues

Follow this link to the Slicer3 bug tracker. Please select the usability issue category when browsing or contributing.

More Information

Acknowledgment

The development of this tool was supported by funding from grants NS050568 and NS40068 from the National Institute of Neurological Disorders and Stroke and grants MH31593, MH40856, from the National Institute of Mental Health.

References