Final Report Summary - DYNAMIC PET/MR (Dynamic PET/MR - New reconstruction methods for Parametric Imaging)
1. Summary of the project objectives
The main project objectives were the development of new methods to improve Positron Emission Tomography (PET) kinetic modelling using Magnetic Resonance Imaging (MRI) information. This objective was divided into two goals: 1) improve input function (IF) estimation and 2) improve Time-Activity curves (TAC) estimation. Furthermore, another additional objective was the simulation of data for evaluation.
2. Description of the work performed since the beginning of the project
The tasks performed in this project consisted mainly on: implementation of Maximum a posterior (MAP) reconstruction methods in an Ordinary Poisson Ordered Subset Expectation Maximisation (OP-OSEM) framework (PRESTO software); Monte-Carlo and analytical simulation of data for validation of methods, with GATE – Geant4 Application for Tomographic Emission and with PRESTO, respectively; evaluation of reconstruction methods and anatomical partial volume correction methods; and development of a new Image-Derived IF (IDIF) method.
3. Description of the achieved results
Estimation of IF and TAC were improved using a MAP reconstruction method (using priors) in conjunction with anatomical partial volume correction (PVC) methods. This resulted in IF and TAC with less noise and partial volume effects. Furthermore, a new image-derived IF method, which used information from dynamic MRI, was also developed. This method improves estimation of IDIF in the early frames, which are the ones with higher noise and more partial volume effects. Furthermore, the use of the MAP reconstruction has been consistently used in clinical datasets with PRESTO software, when the physicians or collaborating researchers require a more detailed reconstruction.
4. Expected final results and their potential impact and use
The expected main results of this research can have great impact in the clinical setting:
1) reduction of patient dose due to improved image quality achieved through reconstruction;
2) more patient comfort due to use of image-derived input function instead of arterial cannulation or even venous blood sampling;
3) improved diagnostic due to higher quality kinetic modelling.
The impact above described is of importance not only on the individual level, but also in terms of healthcare efficiency (time and costs spent on
Moreover, the new IDIF method is also of great interest for the clinical and research community of MR/PET users, which is consistently increasing.
The main project objectives were the development of new methods to improve Positron Emission Tomography (PET) kinetic modelling using Magnetic Resonance Imaging (MRI) information. This objective was divided into two goals: 1) improve input function (IF) estimation and 2) improve Time-Activity curves (TAC) estimation. Furthermore, another additional objective was the simulation of data for evaluation.
2. Description of the work performed since the beginning of the project
The tasks performed in this project consisted mainly on: implementation of Maximum a posterior (MAP) reconstruction methods in an Ordinary Poisson Ordered Subset Expectation Maximisation (OP-OSEM) framework (PRESTO software); Monte-Carlo and analytical simulation of data for validation of methods, with GATE – Geant4 Application for Tomographic Emission and with PRESTO, respectively; evaluation of reconstruction methods and anatomical partial volume correction methods; and development of a new Image-Derived IF (IDIF) method.
3. Description of the achieved results
Estimation of IF and TAC were improved using a MAP reconstruction method (using priors) in conjunction with anatomical partial volume correction (PVC) methods. This resulted in IF and TAC with less noise and partial volume effects. Furthermore, a new image-derived IF method, which used information from dynamic MRI, was also developed. This method improves estimation of IDIF in the early frames, which are the ones with higher noise and more partial volume effects. Furthermore, the use of the MAP reconstruction has been consistently used in clinical datasets with PRESTO software, when the physicians or collaborating researchers require a more detailed reconstruction.
4. Expected final results and their potential impact and use
The expected main results of this research can have great impact in the clinical setting:
1) reduction of patient dose due to improved image quality achieved through reconstruction;
2) more patient comfort due to use of image-derived input function instead of arterial cannulation or even venous blood sampling;
3) improved diagnostic due to higher quality kinetic modelling.
The impact above described is of importance not only on the individual level, but also in terms of healthcare efficiency (time and costs spent on
Moreover, the new IDIF method is also of great interest for the clinical and research community of MR/PET users, which is consistently increasing.