itk::STAPLEImageFilter< TInputImage, TOutputImage > Class Template Reference

The STAPLE filter implements the Simultaneous Truth and Performance Level Estimation algorithm for generating ground truth volumes from a set of binary expert segmentations. More...

#include <itkSTAPLEImageFilter.h>

Inheritance diagram for itk::STAPLEImageFilter< TInputImage, TOutputImage >:

[legend]Collaboration diagram for itk::STAPLEImageFilter< TInputImage, TOutputImage >:

[legend]List of all members.

Public Types
typedef STAPLEImageFilter	Self
typedef ImageToImageFilter< TInputImage, TOutputImage >	Superclass
typedef SmartPointer< Self >	Pointer
typedef SmartPointer< const Self >	ConstPointer
typedef TOutputImage::PixelType	OutputPixelType
typedef TInputImage::PixelType	InputPixelType
typedef NumericTraits< InputPixelType >::RealType	RealType
typedef TInputImage	InputImageType
typedef TOutputImage	OutputImageType
typedef InputImageType::Pointer	InputImagePointer
typedef OutputImageType::Pointer	OutputImagePointer
typedef Superclass::OutputImageRegionType	OutputImageRegionType
Public Member Functions
virtual const char *	GetNameOfClass () const
	itkStaticConstMacro (ImageDimension, unsigned int, TOutputImage::ImageDimension)
const std::vector< double > &	GetSpecificity () const
const std::vector< double > &	GetSensitivity () const
virtual unsigned int	GetElapsedIterations ()
virtual void	SetForegroundValue (InputPixelType _arg)
virtual InputPixelType	GetForegroundValue ()
double	GetSensitivity (unsigned int i)
double	GetSpecificity (unsigned int i)
virtual void	SetMaximumIterations (unsigned int _arg)
virtual unsigned int	GetMaximumIterations ()
virtual void	SetConfidenceWeight (double _arg)
virtual double	GetConfidenceWeight ()
Static Public Member Functions
Pointer	New ()
Protected Member Functions
	STAPLEImageFilter ()
virtual	~STAPLEImageFilter ()
void	GenerateData ()
void	PrintSelf (std::ostream &, Indent) const

---

Detailed Description

template<typename TInputImage, typename TOutputImage>
class itk::STAPLEImageFilter< TInputImage, TOutputImage >

The STAPLE filter implements the Simultaneous Truth and Performance Level Estimation algorithm for generating ground truth volumes from a set of binary expert segmentations.

The STAPLE algorithm treats segmentation as a pixelwise classification, which leads to an averaging scheme that accounts for systematic biases in the behavior of experts in order to generate a fuzzy ground truth volume and simultaneous accuracy assessment of each expert. The ground truth volumes produced by this filter are floating point volumes of values between zero and one that indicate probability of each pixel being in the object targeted by the segmentation.

The STAPLE algorithm is described in

S. Warfield, K. Zou, W. Wells, "Validation of image segmentation and expert quality with an expectation-maximization algorithm" in MICCAI 2002: Fifth International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer-Verlag, Heidelberg, Germany, 2002, pp. 298-306

INPUTS

Input volumes to the STAPLE filter must be binary segmentations of an image, that is, there must be a single foreground value that represents positively classified pixels (pixels that are considered to belong inside the segmentation). Any number of background pixel values may be present in the input images. You can, for example, input volumes with many different labels as long as the structure you are interested in creating ground truth for is consistently labeled among all input volumes. Pixel type of the input volumes does not matter. Specify the label value for positively classified pixels using SetForegroundValue. All other labels will be considered to be negatively classified pixels (background).

Input volumes must all contain the same size RequestedRegions.

OUTPUTS

The STAPLE filter produces a single output volume with a range of floating point values from zero to one. IT IS VERY IMPORTANT TO INSTANTIATE THIS FILTER WITH A FLOATING POINT OUTPUT TYPE (floats or doubles). You may threshold the output above some probability threshold if you wish to produce a binary ground truth.

PARAMETERS

The STAPLE algorithm requires a number of inputs. You may specify any number of input volumes using the SetInput(i, p_i) method, where i ranges from zero to N-1, N is the total number of input segmentations, and p_i is the SmartPointer to the i-th segmentation.

The SetConfidenceWeight parameter is a modifier for the prior probability that any pixel would be classified as inside the target object. This implementation of the STAPLE algorithm automatically calculates prior positive classification probability as the average fraction of the image volume filled by the target object in each input segmentation. The ConfidenceWeight parameter allows for scaling the of this default prior probability: if g_t is the prior probability that a pixel would be classified inside the target object, then g_t is set to g_t * ConfidenceWeight before iterating on the solution. In general ConfidenceWeight should be left to the default of 1.0.

You must provide a foreground value using SetForegroundValue that the STAPLE algorithm will use to identify positively classified pixels in the the input images. All other values in the image will be treated as background values. For example, if your input segmentations consist of 1's everywhere inside the segmented region, then use SetForegroundValue(1).

The STAPLE algorithm is an iterative E-M algorithm and will converge on a solution after some number of iterations that cannot be known a priori. After updating the filter, the total elapsed iterations taken to converge on the solution can be queried through GetElapsedIterations(). You may also specify a MaximumNumberOfIterations, after which the algorithm will stop iterating regardless of whether or not it has converged. This implementation of the STAPLE algorithm will find the solution to within seven digits of precision unless it is stopped early.

Once updated, the Sensitivity (true positive fraction, q) and Specificity (true negative fraction, q) for each expert input volume can be queried using GetSensitivity(i) and GetSpecificity(i), where i is the i-th input volume.

REQUIRED PARAMETERS

The only required parameters for this filter are the ForegroundValue and the input volumes. All other parameters may be safely left to their default values. Please see the paper cited above for more information on the STAPLE algorithm and its parameters. A proper understanding of the algorithm is important for interpreting the results that it produces.

EVENTS

This filter invokes IterationEvent() at each iteration of the E-M algorithm. Setting the AbortGenerateData() flag will cause the algorithm to halt after the current iteration and produce results just as if it had converged. The algorithm makes no attempt to report its progress since the number of iterations needed cannot be known in advance.

Definition at line 119 of file itkSTAPLEImageFilter.h.

---

Member Typedef Documentation

template<typename TInputImage, typename TOutputImage> typedef SmartPointer<const Self> itk::STAPLEImageFilter< TInputImage, TOutputImage >::ConstPointer

Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >. Definition at line 127 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> typedef InputImageType::Pointer itk::STAPLEImageFilter< TInputImage, TOutputImage >::InputImagePointer

Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >. Definition at line 149 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> typedef TInputImage itk::STAPLEImageFilter< TInputImage, TOutputImage >::InputImageType

Image typedef support Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >. Definition at line 147 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> typedef TInputImage::PixelType itk::STAPLEImageFilter< TInputImage, TOutputImage >::InputPixelType

Definition at line 138 of file itkSTAPLEImageFilter.h. Referenced by itk::STAPLEImageFilter< TInputImage, TOutputImage >::~STAPLEImageFilter().

template<typename TInputImage, typename TOutputImage> typedef OutputImageType::Pointer itk::STAPLEImageFilter< TInputImage, TOutputImage >::OutputImagePointer

Reimplemented from itk::ImageSource< TOutputImage >. Definition at line 150 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> typedef Superclass::OutputImageRegionType itk::STAPLEImageFilter< TInputImage, TOutputImage >::OutputImageRegionType

Superclass typedefs. Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >. Definition at line 153 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> typedef TOutputImage itk::STAPLEImageFilter< TInputImage, TOutputImage >::OutputImageType

Some convenient typedefs. Reimplemented from itk::ImageSource< TOutputImage >. Definition at line 148 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> typedef TOutputImage::PixelType itk::STAPLEImageFilter< TInputImage, TOutputImage >::OutputPixelType

Extract some information from the image types. Dimensionality of the two images is assumed to be the same. Definition at line 137 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> typedef SmartPointer<Self> itk::STAPLEImageFilter< TInputImage, TOutputImage >::Pointer

Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >. Definition at line 126 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> typedef NumericTraits<InputPixelType>::RealType itk::STAPLEImageFilter< TInputImage, TOutputImage >::RealType

Definition at line 139 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> typedef STAPLEImageFilter itk::STAPLEImageFilter< TInputImage, TOutputImage >::Self

Standard class typedefs. Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >. Definition at line 124 of file itkSTAPLEImageFilter.h. Referenced by itk::STAPLEImageFilter< TInputImage, TOutputImage >::~STAPLEImageFilter().

template<typename TInputImage, typename TOutputImage> typedef ImageToImageFilter< TInputImage, TOutputImage > itk::STAPLEImageFilter< TInputImage, TOutputImage >::Superclass

Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >. Definition at line 125 of file itkSTAPLEImageFilter.h.

---

Constructor & Destructor Documentation

template<typename TInputImage, typename TOutputImage> itk::STAPLEImageFilter< TInputImage, TOutputImage >::STAPLEImageFilter (   )  [inline, protected]

Definition at line 222 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> virtual itk::STAPLEImageFilter< TInputImage, TOutputImage >::~STAPLEImageFilter (   )  [inline, protected, virtual]

Definition at line 229 of file itkSTAPLEImageFilter.h. References itk::STAPLEImageFilter< TInputImage, TOutputImage >::InputPixelType, and itk::STAPLEImageFilter< TInputImage, TOutputImage >::Self.

---

Member Function Documentation

template<typename TInputImage, typename TOutputImage> void itk::STAPLEImageFilter< TInputImage, TOutputImage >::GenerateData (   )  [protected, virtual]

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. See also: ThreadedGenerateData() Reimplemented from itk::ImageSource< TOutputImage >.

template<typename TInputImage, typename TOutputImage> virtual double itk::STAPLEImageFilter< TInputImage, TOutputImage >::GetConfidenceWeight (   )  [virtual]

Scales the estimated prior probability that a pixel will be inside the targeted object of segmentation. The default prior probability g_t is calculated automatically as the average fraction of positively classified pixels to the total size of the volume (across all input volumes). ConfidenceWeight will scale this default value as g_t = g_t * ConfidenceWeight. In general, ConfidenceWeight should be left to the default of 1.0.

template<typename TInputImage, typename TOutputImage> virtual unsigned int itk::STAPLEImageFilter< TInputImage, TOutputImage >::GetElapsedIterations (   )  [virtual]

Get the number of elapsed iterations of the iterative E-M algorithm.

template<typename TInputImage, typename TOutputImage> virtual InputPixelType itk::STAPLEImageFilter< TInputImage, TOutputImage >::GetForegroundValue (   )  [virtual]

Set get the binary ON value of the input image.

template<typename TInputImage, typename TOutputImage> virtual unsigned int itk::STAPLEImageFilter< TInputImage, TOutputImage >::GetMaximumIterations (   )  [virtual]

Set/Get the maximum number of iterations after which the STAPLE algorithm will be considered to have converged. In general this SHOULD NOT be set and the algorithm should be allowed to converge on its own.

template<typename TInputImage, typename TOutputImage> virtual const char* itk::STAPLEImageFilter< TInputImage, TOutputImage >::GetNameOfClass (   )  const [virtual]

Run-time type information (and related methods) Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >.

template<typename TInputImage, typename TOutputImage> double itk::STAPLEImageFilter< TInputImage, TOutputImage >::GetSensitivity (  unsigned int  i  )  [inline]

After the filter is updated, this method returns the Sensitivity (true positive fraction, p) value for the i-th expert input volume. Definition at line 178 of file itkSTAPLEImageFilter.h. References itkExceptionMacro.

template<typename TInputImage, typename TOutputImage> const std::vector<double>& itk::STAPLEImageFilter< TInputImage, TOutputImage >::GetSensitivity (   )  const [inline]

After the filter is updated, this method returns a std::vector<double> of all Sensitivity (true positive fraction, p) values for the expert input volumes. Definition at line 171 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> double itk::STAPLEImageFilter< TInputImage, TOutputImage >::GetSpecificity (  unsigned int  i  )  [inline]

After the filter is updated, this method returns the Specificity (true negative fraction, q) value for the i-th expert input volume. Definition at line 190 of file itkSTAPLEImageFilter.h. References itkExceptionMacro.

template<typename TInputImage, typename TOutputImage> const std::vector<double>& itk::STAPLEImageFilter< TInputImage, TOutputImage >::GetSpecificity (   )  const [inline]

After the filter is updated, this method returns a std::vector<double> of all Specificity (true negative fraction, q) values for the expert input volumes. Definition at line 163 of file itkSTAPLEImageFilter.h.

template<typename TInputImage, typename TOutputImage> itk::STAPLEImageFilter< TInputImage, TOutputImage >::itkStaticConstMacro (  ImageDimension  , unsigned  int, TOutputImage::ImageDimension  )

Extract some information from the image types. Dimensionality of the two images is assumed to be the same.

template<typename TInputImage, typename TOutputImage> Pointer itk::STAPLEImageFilter< TInputImage, TOutputImage >::New (   )  [static]

Method for creation through the object factory. Reimplemented from itk::Object.

template<typename TInputImage, typename TOutputImage> void itk::STAPLEImageFilter< TInputImage, TOutputImage >::PrintSelf (  std::ostream &  , Indent  )  const [protected, virtual]

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 >.

template<typename TInputImage, typename TOutputImage> virtual void itk::STAPLEImageFilter< TInputImage, TOutputImage >::SetConfidenceWeight (  double  _arg  )  [virtual]

Scales the estimated prior probability that a pixel will be inside the targeted object of segmentation. The default prior probability g_t is calculated automatically as the average fraction of positively classified pixels to the total size of the volume (across all input volumes). ConfidenceWeight will scale this default value as g_t = g_t * ConfidenceWeight. In general, ConfidenceWeight should be left to the default of 1.0.

template<typename TInputImage, typename TOutputImage> virtual void itk::STAPLEImageFilter< TInputImage, TOutputImage >::SetForegroundValue (  InputPixelType  _arg  )  [virtual]

Set get the binary ON value of the input image.

template<typename TInputImage, typename TOutputImage> virtual void itk::STAPLEImageFilter< TInputImage, TOutputImage >::SetMaximumIterations (  unsigned int  _arg  )  [virtual]

Set/Get the maximum number of iterations after which the STAPLE algorithm will be considered to have converged. In general this SHOULD NOT be set and the algorithm should be allowed to converge on its own.

---

The documentation for this class was generated from the following file:

• itkSTAPLEImageFilter.h

---