This class is designed to manage the optimization part of an iterative reconstruction. More...

#include <oOptimizerManager.hh>

Collaboration diagram for oOptimizerManager:

Public Member Functions
	oOptimizerManager ()
	The constructor of oOptimizerManager.
	~oOptimizerManager ()
	The destructor of oOptimizerManager.
int	CheckParameters ()
	A function used to check the parameters settings.
int	Initialize ()
	A function used to initialize the manager and the optimizer it manages.
int	PreDataUpdateStep (int a_iteration, int a_nbIterations, int a_subset, int a_nbSubsets)
	A function that simply calls the eponym function from the vOptimizer.
int	PostDataUpdateStep (int a_iteration, int a_nbIterations, int a_subset, int a_nbSubsets)
	A function that simply calls the eponym function from the vOptimizer.
int	DataUpdateStep (oProjectionLine ap_Line, oImageSpace ap_Image, vEvent *ap_Event, int a_bed, int a_timeFrame, int a_respGate, int a_cardGate, int a_iteration, int a_thread)
	A function dedicated to the update step in the data space (for each event inside the loop)
int	ImageUpdateStep (oImageSpace *ap_Image, int a_iteration, int a_nbSubsets)
	A function dedicated to the update step in the image space (performed after the loop on events)
void	SetVerbose (int a_verboseLevel)
	Set the verbose level.
void	SetOptionsOptimizer (const string &a_optionsOptimizer)
	Set the optimizer projection options contained in the provided string.
void	SetOptionsPenalty (const string &a_optionsPenalty)
	Set the penalty projection options contained in the provided string.
void	SetImageDimensionsAndQuantification (oImageDimensionsAndQuantification *ap_ImageDimensionsAndQuantification)
	Set the image dimensions in use.
void	SetNbTOFBins (int a_nbTOFBins)
	Set the number of TOF bins in use.
void	SetDataMode (int a_dataMode)
	Set the mode of the data (histogram, list-mode)
void	SetDataType (int a_dataType)
	Set the type of the data (pet, spect, etc)
void	SetOptimizerFOMFlag (bool a_optimizerFOMFlag)
	Set the optimizer FOM flag that specifies if some figures-of-merit (FOM) will be computed in the data space.
void	SetOptimizerImageStatFlag (bool a_optimizerImageStatFlag)
	Set the optimizer image stat flag that specifies if some basic statistics about image update is computed.
int	GetNbBackwardImages ()
	Return the number of backward images used by the optimizer, explaining why the eponym function of vOptimizer is called.
FLTNB	GetInitialValue ()
	Return the initial image value used by the optimizer, explaining why the eponym function of vOptimizer is called.
Private Member Functions
int	ParseOptionsAndInitializeOptimizerAndPenalty ()
Private Attributes
oImageDimensionsAndQuantification *	mp_ImageDimensionsAndQuantification
int	m_dataMode
int	m_dataType
int	m_nbTOFBins
string	m_optionsOptimizer
string	m_optionsPenalty
bool	m_optimizerFOMFlag
bool	m_optimizerImageStatFlag
vOptimizer *	mp_Optimizer
vPenalty *	mp_Penalty
int	m_verbose

Detailed Description

This class is designed to manage the optimization part of an iterative reconstruction.

As each manager class, it is created in the main program, all parameters are then set, checked, and the manager is initialized. The manager is then used by the algorithm itself, where the function DataUpdateStep() function is called for each event to apply the forward projection, perform operations in the data space and apply the back-proj, based on a oProjectionLine, a vEvent and a vOptimizer. The ImageUpdateStep() function is called after the loop on events to apply the update operations in the image space, based on the back-projected correction images.

Definition at line 29 of file oOptimizerManager.hh.

Constructor & Destructor Documentation

oOptimizerManager::oOptimizerManager ( )

The constructor of oOptimizerManager.

This is the default and unique constructor. It does not take any parameter and its role is only to affect default values to each member of the class.

Definition at line 27 of file oOptimizerManager.cc.

oOptimizerManager::~oOptimizerManager ( )

The destructor of oOptimizerManager.

This is the default and unique destructor. It does not take any parameter and its role is only to free or delete all structures that were built by this class.

Definition at line 55 of file oOptimizerManager.cc.

Member Function Documentation

int oOptimizerManager::CheckParameters ( )

A function used to check the parameters settings.

This function does not take any parameter and is used to check that all mandatory members were correctly parameterized.

Returns:: An integer reflecting the check status; 0 if no problem, another value otherwise.

Definition at line 64 of file oOptimizerManager.cc.

Here is the caller graph for this function:

int oOptimizerManager::DataUpdateStep	(	oProjectionLine *	ap_Line,
		oImageSpace *	ap_Image,
		vEvent *	ap_Event,
		int	a_bed,
		int	a_timeFrame,
		int	a_respGate,
		int	a_cardGate,
		int	a_iteration,
		int	a_thread
	)

A function dedicated to the update step in the data space (for each event inside the loop)

Parameters:

oProjectionLine*	ap_Line
oImageSpace*	ap_Image
vEvent*	ap_Event
int	a_bed
int	a_timeFrame
int	a_respGate
int	a_cardGate
int	a_iteration
int	a_thread

This function will call many functions of the vOptimizer that split up the update step in the data space in smaller steps: 1. forward projection, 2. optional step, 3. backproj sensitivity for histogram mode, 4. optional step, 5. compute corrections, 6. optional step, 7. backproj corrections, 8. compute FOMs. This function is called by the iterative algorithm, inside the loop on all events.

Returns:: An integer reflecting the execution status; 0 if no problem, another value otherwise.

Definition at line 386 of file oOptimizerManager.cc.

Here is the call graph for this function:

Here is the caller graph for this function:

public inline FLTNB oOptimizerManager::GetInitialValue ( ) [inline]

Return the initial image value used by the optimizer, explaining why the eponym function of vOptimizer is called.

Definition at line 219 of file oOptimizerManager.hh.

Here is the call graph for this function:

Here is the caller graph for this function:

public inline int oOptimizerManager::GetNbBackwardImages ( ) [inline]

Return the number of backward images used by the optimizer, explaining why the eponym function of vOptimizer is called.

Definition at line 213 of file oOptimizerManager.hh.

Here is the call graph for this function:

Here is the caller graph for this function:

int oOptimizerManager::ImageUpdateStep	(	oImageSpace *	ap_Image,
		int	a_iteration,
		int	a_nbSubsets
	)

A function dedicated to the update step in the image space (performed after the loop on events)

Parameters:

oImageSpace*	ap_Image
int	a_iteration
int	a_nbSubsets

This function will update the visited voxels first (see in vOptimizer for details), manage the call for penalty computation and call the image update step function specific to the optimizer.

Returns:: An integer reflecting the execution status; 0 if no problem, another value otherwise.

Definition at line 475 of file oOptimizerManager.cc.

Here is the call graph for this function:

Here is the caller graph for this function:

int oOptimizerManager::Initialize ( )

A function used to initialize the manager and the optimizer it manages.

This function does not take any parameter and is used to initialize everything that should be initialized. In a few words, it parses the options, then creates and initializes the optimizer based on the provided options.

Returns:: An integer reflecting the initialization status; 0 if no problem, another value otherwise.

Definition at line 105 of file oOptimizerManager.cc.

Here is the call graph for this function:

Here is the caller graph for this function:

int oOptimizerManager::ParseOptionsAndInitializeOptimizerAndPenalty ( ) [private]

Definition at line 126 of file oOptimizerManager.cc.

Here is the call graph for this function:

Here is the caller graph for this function:

int oOptimizerManager::PostDataUpdateStep	(	int	a_iteration,
		int	a_nbIterations,
		int	a_subset,
		int	a_nbSubsets
	)

A function that simply calls the eponym function from the vOptimizer.

Parameters:

int	a_iteration
int	a_nbIterations
int	a_subset
int	a_nbSubsets

Returns:: An integer reflecting the execution status; 0 if no problem, another value otherwise.

Definition at line 369 of file oOptimizerManager.cc.

Here is the call graph for this function:

Here is the caller graph for this function:

int oOptimizerManager::PreDataUpdateStep	(	int	a_iteration,
		int	a_nbIterations,
		int	a_subset,
		int	a_nbSubsets
	)

A function that simply calls the eponym function from the vOptimizer.

Parameters:

int	a_iteration
int	a_nbIterations
int	a_subset
int	a_nbSubsets

Returns:: An integer reflecting the execution status; 0 if no problem, another value otherwise.

Definition at line 352 of file oOptimizerManager.cc.

Here is the call graph for this function:

Here is the caller graph for this function:

public inline void oOptimizerManager::SetDataMode ( int a_dataMode ) [inline]

Set the mode of the data (histogram, list-mode)

Parameters:

int	a_dataMode

Definition at line 186 of file oOptimizerManager.hh.

Here is the caller graph for this function:

public inline void oOptimizerManager::SetDataType ( int a_dataType ) [inline]

Set the type of the data (pet, spect, etc)

Parameters:

int	a_dataType

Definition at line 193 of file oOptimizerManager.hh.

Here is the caller graph for this function:

public inline void oOptimizerManager::SetImageDimensionsAndQuantification ( oImageDimensionsAndQuantification * ap_ImageDimensionsAndQuantification ) [inline]

Set the image dimensions in use.

Parameters:

oImageDimensionsAndQuantification* ap_ImageDimensionsAndQuantification

Definition at line 172 of file oOptimizerManager.hh.

Here is the caller graph for this function:

public inline void oOptimizerManager::SetNbTOFBins ( int a_nbTOFBins ) [inline]

Set the number of TOF bins in use.

Parameters:

int	a_nbTOFBins

Definition at line 179 of file oOptimizerManager.hh.

Here is the caller graph for this function:

public inline void oOptimizerManager::SetOptimizerFOMFlag ( bool a_optimizerFOMFlag ) [inline]

Set the optimizer FOM flag that specifies if some figures-of-merit (FOM) will be computed in the data space.

Parameters:

bool	a_optimizerFOMFlag

Definition at line 200 of file oOptimizerManager.hh.

Here is the caller graph for this function:

public inline void oOptimizerManager::SetOptimizerImageStatFlag ( bool a_optimizerImageStatFlag ) [inline]

Set the optimizer image stat flag that specifies if some basic statistics about image update is computed.

Parameters:

bool	a_optimizerImageStatFlag

Definition at line 207 of file oOptimizerManager.hh.

Here is the caller graph for this function:

public inline void oOptimizerManager::SetOptionsOptimizer ( const string & a_optionsOptimizer ) [inline]

Set the optimizer projection options contained in the provided string.

Parameters:

const string& a_optionsOptimizer

Definition at line 158 of file oOptimizerManager.hh.

Here is the caller graph for this function:

public inline void oOptimizerManager::SetOptionsPenalty ( const string & a_optionsPenalty ) [inline]

Set the penalty projection options contained in the provided string.

Parameters:

const string& a_optionsPenalty

Definition at line 165 of file oOptimizerManager.hh.

Here is the caller graph for this function:

public inline void oOptimizerManager::SetVerbose ( int a_verboseLevel ) [inline]

Set the verbose level.

Parameters:

int	a_verboseLevel

Definition at line 151 of file oOptimizerManager.hh.

Here is the caller graph for this function:

Member Data Documentation

int oOptimizerManager::m_dataMode [private]

Flag indicating if the data is List (=0) or Histogram (=1) mode

Definition at line 228 of file oOptimizerManager.hh.

int oOptimizerManager::m_dataType [private]

Flag indicating if the data is PET (=0),SPECT (=1) or TRANSMISSION type (=2)

Definition at line 229 of file oOptimizerManager.hh.

int oOptimizerManager::m_nbTOFBins [private]

The number of TOF bins in use

Definition at line 230 of file oOptimizerManager.hh.

bool oOptimizerManager::m_optimizerFOMFlag [private]

Flag that says if some figures-of-merit will be computed in the data space

Definition at line 233 of file oOptimizerManager.hh.

bool oOptimizerManager::m_optimizerImageStatFlag [private]

Flag that says if some basic statistics about the image update will be computed

Definition at line 234 of file oOptimizerManager.hh.

string oOptimizerManager::m_optionsOptimizer [private]

The string containing options for the optimizer projections

Definition at line 231 of file oOptimizerManager.hh.

string oOptimizerManager::m_optionsPenalty [private]

The string containing options for the penalty projections

Definition at line 232 of file oOptimizerManager.hh.

int oOptimizerManager::m_verbose [private]

The verbose level

Definition at line 237 of file oOptimizerManager.hh.

oImageDimensionsAndQuantification* oOptimizerManager::mp_ImageDimensionsAndQuantification [private]

Pointer to the oImageDimensionsAndQuantification object in use

Definition at line 227 of file oOptimizerManager.hh.

vOptimizer* oOptimizerManager::mp_Optimizer [private]

The actual optimizer in use

Definition at line 235 of file oOptimizerManager.hh.

vPenalty* oOptimizerManager::mp_Penalty [private]

The actual penalty in use (optional)

Definition at line 236 of file oOptimizerManager.hh.

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

Public Member Functions

Private Member Functions

Private Attributes

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation

Member Data Documentation