ITK
4.13.0
Insight Segmentation and Registration Toolkit
|
#include <itkN4BiasFieldCorrectionImageFilter.h>
Implementation of the N4 bias field correction algorithm.
The nonparametric nonuniform intensity normalization (N3) algorithm, as introduced by Sled et al. in 1998 is a method for correcting nonuniformity associated with MR images. The algorithm assumes a simple parametric model (Gaussian) for the bias field and does not require tissue class segmentation. In addition, there are only a couple of parameters to tune with the default values performing quite well. N3 has been publicly available as a set of perl scripts (http://www.bic.mni.mcgill.ca/ServicesSoftwareAdvancedImageProcessingTools/HomePage)
The N4 algorithm, encapsulated with this class, is a variation of the original N3 algorithm with the additional benefits of an improved B-spline fitting routine which allows for multiple resolutions to be used during the correction process. We also modify the iterative update component of algorithm such that the residual bias field is continually updated
Notes for the user:
The basic algorithm iterates between sharpening the intensity histogram of the corrected input image and spatially smoothing those results with a B-spline scalar field estimate of the bias field.
Contributed by Nicholas J. Tustison, James C. Gee in the Insight Journal paper: https://hdl.handle.net/10380/3053
J.G. Sled, A.P. Zijdenbos and A.C. Evans. "A Nonparametric Method for Automatic Correction of Intensity Nonuniformity in Data" IEEE Transactions on Medical Imaging, Vol 17, No 1. Feb 1998.
N.J. Tustison, B.B. Avants, P.A. Cook, Y. Zheng, A. Egan, P.A. Yushkevich, and J.C. Gee. "N4ITK: Improved N3 Bias Correction" IEEE Transactions on Medical Imaging, 29(6):1310-1320, June 2010.
Definition at line 95 of file itkN4BiasFieldCorrectionImageFilter.h.
Static Public Member Functions | |
static Pointer | New () |
Static Public Member Functions inherited from itk::Object | |
static bool | GetGlobalWarningDisplay () |
static void | GlobalWarningDisplayOff () |
static void | GlobalWarningDisplayOn () |
static Pointer | New () |
static void | SetGlobalWarningDisplay (bool flag) |
Static Public Member Functions inherited from itk::LightObject | |
static void | BreakOnError () |
static Pointer | New () |
Static Public Attributes | |
static const unsigned int | ImageDimension = TInputImage::ImageDimension |
Static Public Attributes inherited from itk::ImageToImageFilter< TInputImage, TOutputImage > | |
static const unsigned int | InputImageDimension = TInputImage::ImageDimension |
static const unsigned int | OutputImageDimension = TOutputImage::ImageDimension |
Static Public Attributes inherited from itk::ImageSource< TOutputImage > | |
static const unsigned int | OutputImageDimension = TOutputImage::ImageDimension |
Private Member Functions | |
RealType | CalculateConvergenceMeasurement (const RealImageType *, const RealImageType *) const |
RealImagePointer | ReconstructBiasField (BiasFieldControlPointLatticeType *) |
RealImagePointer | SharpenImage (const RealImageType *) const |
RealImagePointer | UpdateBiasFieldEstimate (RealImageType *) |
Additional Inherited Members | |
Protected Types inherited from itk::ImageToImageFilter< TInputImage, TOutputImage > | |
typedef ImageToImageFilterDetail::ImageRegionCopier < itkGetStaticConstMacro(OutputImageDimension), itkGetStaticConstMacro(InputImageDimension) > | InputToOutputRegionCopierType |
typedef ImageToImageFilterDetail::ImageRegionCopier < itkGetStaticConstMacro(InputImageDimension), itkGetStaticConstMacro(OutputImageDimension) > | OutputToInputRegionCopierType |
Static Protected Member Functions inherited from itk::ImageSource< TOutputImage > | |
static const ImageRegionSplitterBase * | GetGlobalDefaultSplitter () |
static ITK_THREAD_RETURN_TYPE | ThreaderCallback (void *arg) |
Protected Attributes inherited from itk::ProcessObject | |
TimeStamp | m_OutputInformationMTime |
bool | m_Updating |
Protected Attributes inherited from itk::LightObject | |
AtomicInt< int > | m_ReferenceCount |
typedef BSplineFilterType::ArrayType itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::ArrayType |
Definition at line 138 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef BSplineFilterType::PointDataImageType itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::BiasFieldControlPointLatticeType |
Definition at line 137 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef BSplineScatteredDataPointSetToImageFilter< PointSetType, ScalarImageType> itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::BSplineFilterType |
B-sline filter typedefs
Definition at line 136 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef SmartPointer<const Self> itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::ConstPointer |
Definition at line 103 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef TInputImage itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::InputImageType |
Some convenient typedefs.
Definition at line 116 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef TMaskImage itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::MaskImageType |
Definition at line 118 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef MaskImageType::PixelType itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::MaskPixelType |
Definition at line 119 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef TOutputImage itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::OutputImageType |
Definition at line 117 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef SmartPointer<Self> itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::Pointer |
Definition at line 102 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef PointSetType::Pointer itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::PointSetPointer |
B-spline smoothing filter argument typedefs
Definition at line 130 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef PointSet<ScalarType, itkGetStaticConstMacro( ImageDimension )> itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::PointSetType |
B-spline smoothing filter argument typedefs
Definition at line 128 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef PointSetType::PointType itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::PointType |
B-spline smoothing filter argument typedefs
Definition at line 131 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef RealImageType::Pointer itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::RealImagePointer |
Definition at line 123 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef Image<RealType, ImageDimension> itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::RealImageType |
Definition at line 122 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef float itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::RealType |
Definition at line 121 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef Image<ScalarType, itkGetStaticConstMacro( ImageDimension )> itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::ScalarImageType |
B-spline smoothing filter argument typedefs
Definition at line 129 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef Vector<RealType, 1> itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::ScalarType |
B-spline smoothing filter argument typedefs
Definition at line 127 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef N4BiasFieldCorrectionImageFilter itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::Self |
Standard class typedefs.
Definition at line 100 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef ImageToImageFilter<TInputImage, TOutputImage> itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::Superclass |
Definition at line 101 of file itkN4BiasFieldCorrectionImageFilter.h.
typedef Array<unsigned int> itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::VariableSizeArrayType |
Definition at line 124 of file itkN4BiasFieldCorrectionImageFilter.h.
|
protected |
|
inlineoverrideprotected |
Definition at line 383 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Convergence is determined by the coefficient of variation of the difference image between the current bias field estimate and the previous estimate.
|
virtual |
Create an object from an instance, potentially deferring to a factory. This method allows you to create an instance of an object that is exactly the same type as the referring object. This is useful in cases where an object has been cast back to a base class.
Reimplemented from itk::Object.
|
overrideprotectedvirtual |
A version of GenerateData() specific for image processing filters. This implementation will split the processing across multiple threads. The buffer is allocated by this method. Then the BeforeThreadedGenerateData() method is called (if provided). Then, a series of threads are spawned each calling ThreadedGenerateData(). After all the threads have completed processing, the AfterThreadedGenerateData() method is called (if provided). If an image processing filter cannot be threaded, the filter should provide an implementation of GenerateData(). That implementation is responsible for allocating the output buffer. If a filter an be threaded, it should NOT provide a GenerateData() method but should provide a ThreadedGenerateData() instead.
Reimplemented from itk::ImageSource< TOutputImage >.
|
virtual |
Get the full width at half maximum parameter characterizing the width of the Gaussian deconvolution. Default = 0.15.
|
inline |
Get confidence image function. If a confidence image is specified, estimation of the bias field weights the contribution of each voxel according the value of the corresponding voxel in the confidence image. For example, when estimating the bias field using brain , one can use a soft segmentation of the white matter as the confidence image instead of using a hard segmentation of the white matter as the mask image (as has been done in the literature) as an alternative strategy to estimating the bias field.
Definition at line 218 of file itkN4BiasFieldCorrectionImageFilter.h.
References itk::ProcessObject::GetInput().
|
virtual |
Get the convergence threshold. Convergence is determined by the coefficient of variation of the difference image between the current bias field estimate and the previous estimate. If this value is less than the specified threshold, the algorithm proceeds to the next fitting level or terminates if it is at the last level.
|
virtual |
Get the current convergence measurement. This is a helper function for reporting observations.
|
virtual |
Get the current fitting level. This is a helper function for reporting observations.
|
virtual |
Get the number of elapsed iterations. This is a helper function for reporting observations.
|
virtual |
Typically, a reduced size image is used as input to the N4 filter using something like itkShrinkImageFilter. Since the output is a corrected version of the input, the user will probably want to apply the bias field correction to the full resolution image. This can be done by using the LogBiasFieldControlPointLattice to reconstruct the bias field at the full image resolution (using the class BSplineControlPointImageFilter).
|
inline |
Get mask image function. If a binary mask image is specified, only those input image voxels inside the mask image values are used in estimating the bias field.
Definition at line 163 of file itkN4BiasFieldCorrectionImageFilter.h.
References itk::ProcessObject::GetInput().
|
virtual |
|
virtual |
Get the maximum number of iterations specified at each fitting level. Default = 50.
|
virtual |
Runtime information support.
Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >.
|
virtual |
Get the control point grid size defining the B-spline estimate of the scalar bias field. In each dimension, the B-spline mesh size is equal to the number of control points in that dimension minus the spline order. Default = 4 control points in each dimension for a mesh size of 1 in each dimension.
|
virtual |
Get the number of fitting levels. One of the contributions of N4 is the introduction of a multi-scale approach to fitting. This allows one to specify a B-spline mesh size for initial fitting followed by a doubling of the mesh resolution for each subsequent fitting level. Default = 1 level.
|
virtual |
Get number of bins defining the log input intensity histogram. Default = 200.
|
virtual |
Get the spline order defining the bias field estimate. Default = 3.
|
virtual |
Use a mask label for identifying mask functionality. See SetMaskLabel. Defaults to true.
|
virtual |
Get the noise estimate defining the Wiener filter. Default = 0.01.
|
static |
Standard New method.
|
overrideprotectedvirtual |
Methods invoked by Print() to print information about the object including superclasses. Typically not called by the user (use Print() instead) but used in the hierarchical print process to combine the output of several classes.
Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >.
|
private |
Reconstruct bias field given the control point lattice.
|
virtual |
Set the full width at half maximum parameter characterizing the width of the Gaussian deconvolution. Default = 0.15.
|
inline |
Set confidence image function. If a confidence image is specified, estimation of the bias field weights the contribution of each voxel according the value of the corresponding voxel in the confidence image. For example, when estimating the bias field using brain , one can use a soft segmentation of the white matter as the confidence image instead of using a hard segmentation of the white matter as the mask image (as has been done in the literature) as an alternative strategy to estimating the bias field.
Definition at line 201 of file itkN4BiasFieldCorrectionImageFilter.h.
References itk::ProcessObject::SetNthInput().
Referenced by itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::SetInput3().
|
virtual |
Set the convergence threshold. Convergence is determined by the coefficient of variation of the difference image between the current bias field estimate and the previous estimate. If this value is less than the specified threshold, the algorithm proceeds to the next fitting level or terminates if it is at the last level.
|
inline |
The image expected for input for bias correction.
Definition at line 143 of file itkN4BiasFieldCorrectionImageFilter.h.
References itk::ImageToImageFilter< TInputImage, TOutputImage >::SetInput().
|
inline |
Set mask image function. If a binary mask image is specified, only those input image voxels inside the mask image values are used in estimating the bias field.
Definition at line 155 of file itkN4BiasFieldCorrectionImageFilter.h.
References itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::SetMaskImage().
|
inline |
Set confidence image function. If a confidence image is specified, estimation of the bias field weights the contribution of each voxel according the value of the corresponding voxel in the confidence image. For example, when estimating the bias field using brain , one can use a soft segmentation of the white matter as the confidence image instead of using a hard segmentation of the white matter as the mask image (as has been done in the literature) as an alternative strategy to estimating the bias field.
Definition at line 205 of file itkN4BiasFieldCorrectionImageFilter.h.
References itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::SetConfidenceImage().
|
inline |
Set mask image function. If a binary mask image is specified, only those input image voxels inside the mask image values are used in estimating the bias field.
Definition at line 151 of file itkN4BiasFieldCorrectionImageFilter.h.
References itk::ProcessObject::SetNthInput().
Referenced by itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::SetInput2().
|
virtual |
|
virtual |
Set the maximum number of iterations specified at each fitting level. Default = 50.
|
virtual |
Set the control point grid size defining the B-spline estimate of the scalar bias field. In each dimension, the B-spline mesh size is equal to the number of control points in that dimension minus the spline order. Default = 4 control points in each dimension for a mesh size of 1 in each dimension.
|
virtual |
Set the number of fitting levels. One of the contributions of N4 is the introduction of a multi-scale approach to fitting. This allows one to specify a B-spline mesh size for initial fitting followed by a doubling of the mesh resolution for each subsequent fitting level. Default = 1 level.
Referenced by itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::SetNumberOfFittingLevels().
|
inline |
Set the number of fitting levels. One of the contributions of N4 is the introduction of a multi-scale approach to fitting. This allows one to specify a B-spline mesh size for initial fitting followed by a doubling of the mesh resolution for each subsequent fitting level. Default = 1 level.
Definition at line 306 of file itkN4BiasFieldCorrectionImageFilter.h.
References itk::FixedArray< TValue, VLength >::Fill(), and itk::N4BiasFieldCorrectionImageFilter< TInputImage, TMaskImage, TOutputImage >::SetNumberOfFittingLevels().
|
virtual |
Set number of bins defining the log input intensity histogram. Default = 200.
|
virtual |
Set the spline order defining the bias field estimate. Default = 3.
|
virtual |
Use a mask label for identifying mask functionality. See SetMaskLabel. Defaults to true.
|
virtual |
Set the noise estimate defining the Wiener filter. Default = 0.01.
|
private |
Sharpen the intensity histogram of the current estimate of the corrected image and map those results to a new estimate of the unsmoothed corrected image.
|
private |
Given the unsmoothed estimate of the bias field, this function smooths the estimate and adds the resulting control point values to the total bias field estimate.
|
virtual |
Use a mask label for identifying mask functionality. See SetMaskLabel. Defaults to true.
|
virtual |
Use a mask label for identifying mask functionality. See SetMaskLabel. Defaults to true.
|
static |
ImageDimension constants
Definition at line 113 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 433 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 439 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 440 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 441 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 438 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 446 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 425 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 437 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 449 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 450 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 431 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 448 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 426 of file itkN4BiasFieldCorrectionImageFilter.h.
|
private |
Definition at line 432 of file itkN4BiasFieldCorrectionImageFilter.h.