A helper class to store the model matrix for a linear kinetic model. More...

#include "stir/modelling/ModelMatrix.h"

Inheritance diagram for stir::ModelMatrix< num_param >:

Public Member Functions
	ModelMatrix ()
	default constructor

	~ModelMatrix ()
	default destructor

void	read_from_file (const std::string input_string)
	Implementation to read the model matrix. More...

Succeeded	write_to_file (const std::string output_string)
	Implementation to write the model matrix to a text file. More...

void	threshold_model_array (const float threshold_value)
	Function to give the threshold_value to the all elements of the model_array which lower value than the threshold_value.

void	uncalibrate (const float cal_factor)

void	scale_model_matrix (const float scale_factor)

void	convert_to_total_frame_counts (const TimeFrameDefinitions &time_frame_definitions)

Functions to get parameters @{
Array< 2, float >	get_model_array () const

const VectorWithOffset< float >	get_model_array_sum () const

VectorWithOffset< float >	get_time_vector () const

Functions to set parameters @{
void	set_model_array (const Array< 2, float > &model_array)

void	set_time_vector (const VectorWithOffset< float > &time_vector)

void	set_is_uncalibrated (const bool is_uncalibrated)
	Function to set _is_calibrated boolean true or false.

void	set_is_in_correct_scale (const bool in_correct_scale)


void	multiply_dynamic_image_with_model_and_add_to_input (ParametricVoxelsOnCartesianGrid &parametric_image, const DynamicDiscretisedDensity &dynamic_image) const
	multiply (transpose) model-matrix with dynamic image and add result to original `parametric_image` More...

void	multiply_dynamic_image_with_model (ParametricVoxelsOnCartesianGrid &parametric_image, const DynamicDiscretisedDensity &dynamic_image) const
	multiply (transpose) model-matrix with dynamic image (overwriting original content of `parametric_image`) More...

void	multiply_parametric_image_with_model_and_add_to_input (DynamicDiscretisedDensity &dynamic_image, const ParametricVoxelsOnCartesianGrid &parametric_image) const
	multiply model-matrix with parametric image and add result to original `dynamic_image`

void	multiply_parametric_image_with_model (DynamicDiscretisedDensity &dynamic_image, const ParametricVoxelsOnCartesianGrid &parametric_image) const
	multiply model-matrix with parametric image (overwriting original content of `dynamic_image`) More...

void	normalise_parametric_image_with_model_sum (ParametricVoxelsOnCartesianGrid &parametric_image_out, const ParametricVoxelsOnCartesianGrid &parametric_image) const

Detailed Description

template<int num_param>
class stir::ModelMatrix< num_param >

A helper class to store the model matrix for a linear kinetic model.

Member Function Documentation

◆ read_from_file()

template<int num_param>

void stir::ModelMatrix< num_param >::read_from_file ( const std::string input_string )

inline

Implementation to read the model matrix.

Implementation to read the model matrix from a text file

Warning: In this way the information about the calibration _is_uncalibrated and the counts _is_converted is not passed.

Referenced by stir::modellingTests::run_tests().

◆ write_to_file()

template<int num_param>

Succeeded stir::ModelMatrix< num_param >::write_to_file ( const std::string output_string )

inline

Implementation to write the model matrix to a text file.

Implementation to write the model matrix.

Referenced by stir::modellingTests::run_tests().

◆ uncalibrate()

template<int num_param>

void stir::ModelMatrix< num_param >::uncalibrate ( const float cal_factor )

inline

Function to divide with the calibration factor the model array. Calibrated ModelMatrix means that the counts are in kBq/ml, while uncalibrated means that it will be to the same units as the reconstructed images.

◆ scale_model_matrix()

template<int num_param>

void stir::ModelMatrix< num_param >::scale_model_matrix ( const float scale_factor )

inline

Function to multiply with the scale factor the model array. Scaled ModelMatrix means that the counts are already scaled to the correct, while not scaled means that it needs to be scaled.

◆ convert_to_total_frame_counts()

template<int num_param>

void stir::ModelMatrix< num_param >::convert_to_total_frame_counts ( const TimeFrameDefinitions & time_frame_definitions )

inline

Multiply with the duration to convert the count rate to total counts in the time frame. Converted ModelMatrix means that it is in total counts in respect to the time_frame_duration, while not converted sets the _is_converted to false and means that it will be in "mean count rate".

◆ multiply_dynamic_image_with_model_and_add_to_input()

template<int num_param>

void stir::ModelMatrix< num_param >::multiply_dynamic_image_with_model_and_add_to_input	(	ParametricVoxelsOnCartesianGrid &	parametric_image,
		const DynamicDiscretisedDensity &	dynamic_image
	)		const

inline

multiply (transpose) model-matrix with dynamic image and add result to original parametric_image

Multiplications of the model with the dynamic or the parametric images. /todo Maybe it will be better to lie in a linear models class.

◆ multiply_dynamic_image_with_model()

template<int num_param>

void stir::ModelMatrix< num_param >::multiply_dynamic_image_with_model	(	ParametricVoxelsOnCartesianGrid &	parametric_image,
		const DynamicDiscretisedDensity &	dynamic_image
	)		const

inline

multiply (transpose) model-matrix with dynamic image (overwriting original content of parametric_image)

Todo:: current implementation first fills first argument with 0 and then calls multiply_dynamic_image_with_model_and_add_to_input(). This is somewhat inefficient.

◆ multiply_parametric_image_with_model()

template<int num_param>

void stir::ModelMatrix< num_param >::multiply_parametric_image_with_model	(	DynamicDiscretisedDensity &	dynamic_image,
		const ParametricVoxelsOnCartesianGrid &	parametric_image
	)		const

inline

multiply model-matrix with parametric image (overwriting original content of dynamic_image)

Todo:: current implementation first fills first argument with 0 and then calls multiply_dynamic_image_with_model_and_add_to_input(). This is somewhat inefficient.

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

/home/sirfuser/devel/STIRdistrib/STIR/src/include/stir/modelling/ModelMatrix.h
/home/sirfuser/devel/STIRdistrib/STIR/src/include/stir/modelling/ModelMatrix.inl

Public Member Functions

Detailed Description

template<int num_param> class stir::ModelMatrix< num_param >

Member Function Documentation

◆ read_from_file()

◆ write_to_file()

◆ uncalibrate()

◆ scale_model_matrix()

◆ convert_to_total_frame_counts()

◆ multiply_dynamic_image_with_model_and_add_to_input()

◆ multiply_dynamic_image_with_model()

◆ multiply_parametric_image_with_model()

template<int num_param>
class stir::ModelMatrix< num_param >