Related Experiment Video
Updated: May 18, 2026
MRI and PET in Mouse Models of Myocardial Infarction
Published on: December 19, 2013
Markov random field and Gaussian mixture for segmented MRI-based partial volume correction in PET
Alexandre Bousse1, Stefano Pedemonte, Benjamin A Thomas
1Institute of Nuclear Medicine, University College London, London NW1 2BU, UK. a.bousse@ucl.ac.uk
Abstract:
In this paper we propose a segmented magnetic resonance imaging (MRI) prior-based maximum penalized likelihood deconvolution technique for positron emission tomography (PET) images. The model assumes the existence of activity classes that behave like a hidden Markov random field (MRF) driven by the segmented MRI. We utilize a mean field approximation to compute the likelihood of the MRF. We tested our method on both simulated and clinical data (brain PET) and compared our results with PET images corrected with the re-blurred Van Cittert (VC) algorithm, the simplified Guven (SG) algorithm and the region-based voxel-wise (RBV) technique. We demonstrated our algorithm outperforms the VC algorithm and outperforms SG and RBV corrections when the segmented MRI is inconsistent (e.g. mis-segmentation, lesions, etc) with the PET image.
Insights
We developed a new method using segmented magnetic resonance imaging (MRI) to improve positron emission tomography (PET) image deconvolution. This technique enhances image quality, especially when MRI and PET data have discrepancies.
Area of Science:
- Medical Imaging
- Image Processing
- Nuclear Medicine
Background:
- Positron emission tomography (PET) imaging is crucial for disease diagnosis and monitoring.
- Image deconvolution techniques are essential for improving PET image resolution and accuracy.
- Accurate anatomical information from magnetic resonance imaging (MRI) can aid PET image analysis.
Purpose of the Study:
- To introduce a novel segmented MRI-prior-based maximum penalized likelihood deconvolution technique for PET imaging.
- To enhance the quality of PET images by incorporating structural information from segmented MRI.
- To evaluate the performance of the proposed method against existing deconvolution algorithms.
Main Methods:
- A maximum penalized likelihood deconvolution technique incorporating segmented MRI priors was developed.
- The method models activity classes as a hidden Markov random field (MRF) driven by segmented MRI.
- A mean field approximation was used to compute the MRF likelihood.
Main Results:
- The proposed method was tested on simulated and clinical brain PET data.
- Performance was compared against the re-blurred Van Cittert (VC), simplified Guven (SG), and region-based voxel-wise (RBV) techniques.
- The new algorithm demonstrated superior performance compared to the VC algorithm.
- It outperformed SG and RBV corrections when MRI segmentation was inconsistent with PET data.
Conclusions:
- The segmented MRI-prior-based deconvolution technique offers improved PET image quality.
- This method is particularly advantageous in cases of MRI-PET data misalignment or mis-segmentation.
- The approach holds promise for more accurate diagnostic interpretations in nuclear medicine.
