https://doi.org/10.1051/epjconf/202328806003
An Extended List-Mode MLEM Algorithm for 3D Compton Image Reconstruction from Multi-View Data
1 Computer Science and Digital Society laboratory, Troyes University of Technology, France
2 Damavan Imaging, 2 rue Gustave Eiffel, 10430 Rosières-Près-Troyes, France
Published online: 21 November 2023
The lack of parallax in the measurements is a big challenge in 3D image reconstruction for handheld Compton cameras. Our solution to this issue is to extend the conventional list-mode maximum-likelihood expectation-maximization (LM-MLEM) 3D reconstruction algorithm to allow the simultaneous use of multi-view Compton data seeking parallax improvement. It involves building a new list-mode simultaneous data space from multi-view Compton events, formulating the associated probabilistic models for the system response matrix and sensitivity, and developing an extended LM-MLEM algorithm. For the performance assessment, we experiment the extended 3D reconstruction algorithm on real multi-view data conducted with a handheld CeBr3 Compton camera developed by Damavan Imaging and a punctual 0.2 (MBq) 22Na source. Various comparative studies with different view numbers, source locations and energy ranges confirm the outperformances of our extended LM-MLEM algorithm.
Key words: 3D reconstruction / handheld Compton camera / LM-MLEM algorithm / multi-view Compton events / real data
© The Authors, published by EDP Sciences, 2023
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
