One-step iterative reconstruction approach based on eigentissue decomposition for spectral photon-counting computed tomography

• DOI: 10.1117/1.jmi.9.4.044003
• Publication Date: 2022-07-27T19:07:41Z
• Authors: Mikaël Simard; Hugo Bouchard
• Keywords: 03 medical and health sciences; 0302 clinical medicine; quantitative imaging; one-step reconstruction; spectral photon-counting computed tomography; eigentissue decomposition; tissue characterization; radiotherapy

Purpose: We propose a one-step tissue characterization method for spectral photon-counting computed tomography (SPCCT) using eigentissue decomposition (ETD), tailored highly accurate human in radiotherapy. Methods: The approach combines Poisson likelihood, spatial prior, and quantitative prior constraining fractions based on expected values tabulated tissues. There are two regularization parameters: α the β prior. is validated realistic simulation environment SPCCT. impact of evaluated virtual phantom. framework tested patient compared with sinogram-based two-step methods [using respectively filtered backprojection (FBP) an iterative second step] post-reconstruction same All use ETD. Results: Optimal performance respect to bias or RMSE achieved different combinations cylindrical Evaluated tissues patient, outperforms approaches quantify proton-stopping power (SPR). mean absolute SPR 0.6% (two-step FBP), iterative), (post-reconstruction), 0.2% (one-step optimized low bias). Following order, 13.3%, 2.5%, 3.2%, 1.5%. Conclusions: Accurate precise ETD can be noisy SPCCT data without need rely methods. more than characterization.