Dose volume histogram-based optimization of image reconstruction parameters for quantitative ⁹⁰ Y-PET imaging

Area of Science:

• Nuclear Medicine
• Radiotherapy Physics
• Medical Imaging

Background:

• Yttrium-90 (⁹⁰Y) microsphere radioembolization is a key treatment for unresectable tumors.
• Volumetric 3D ⁹⁰Y dosimetry using PET imaging is emerging for dose-response studies.
• Current ⁹⁰Y-PET reconstruction optimization relies on recovery coefficient (RC), which is suboptimal for dosimetry.

Purpose of the Study:

• To optimize ⁹⁰Y-PET image reconstruction by minimizing dose volume histogram (DVH) errors for improved volumetric dosimetry.
• To jointly optimize the number of equivalent iterations and post-reconstruction filtration (FWHM) for accurate voxel-level ⁹⁰Y dosimetry.

Main Methods:

• A modified NEMA IEC phantom was used to simulate clinical ⁹⁰Y-PET imaging conditions.
• PET data were acquired and rebinned to various durations, and DVH errors were calculated as RMSE.
• The proposed DVH-based optimization was compared against the conventional mean activity concentration RC.

Main Results:

• Joint optimization of 36 equivalent iterations and 5.2-mm filtration significantly decreased systematic errors in volumetric ⁹⁰Y dosimetry.
• The proposed DVH objective function demonstrated superiority over RC-based optimization for absorbed dose quantification.
• OSEM reconstruction with TOF and PSF modeling was used, showing improved accuracy with the new optimization.

Conclusions:

• Minimizing DVH errors provides a superior objective function for optimizing ⁹⁰Y-PET reconstruction for volumetric dosimetry in therapy.
• The proposed joint optimization of iterations and filtration enhances the accuracy of PET-based ⁹⁰Y dose quantification.
• This DVH-based optimization approach is applicable to various imaging modalities for voxel-level quantitative assessments.