Related Experiment Video
Updated: Jun 17, 2026
Positron Emission Tomography-based Dose Painting Radiation Therapy in a Glioblastoma Rat Model using the Small Animal Radiation Research Platform
Published on: March 24, 2022
A method for partial volume correction of PET-imaged tumor heterogeneity using expectation maximization with a
David L Barbee1, Ryan T Flynn, James E Holden
1Department of Medical Physics, University of Wisconsin School of Medicine and Public Health, 1111 Highland Ave, Madison, WI 53705, USA.
This study introduces a spatially varying partial volume correction (SV-PVC) method to improve positron emission tomography (PET) imaging of heterogeneous tumors. The SV-PVC technique enhances quantitative accuracy by accounting for partial volume effects, crucial for treatment planning and assessment.
Area of Science:
- Medical Imaging
- Nuclear Medicine
- Image Processing
Background:
- Tumor heterogeneities in positron emission tomography (PET) imaging are often affected by partial volume effects.
- These effects can compromise treatment prognosis, assessment, and biologically optimized treatment planning (dose painting).
Purpose of the Study:
- To present a novel method for partial volume correction (PVC) of PET-imaged heterogeneous tumors.
- To develop and validate a spatially varying partial volume correction (SV-PVC) technique.
Main Methods:
- Scanned a point source to obtain spatially varying point spread functions (PSFs).
- Developed continuous expressions for Gaussian widths as a function of radial distance to generate system PSFs.
- Implemented a spatially varying partial volume correction (SV-PVC) technique using expectation maximization (EM) with an automated stopping criterion.
Main Results:
- SV-PVC increased maximum observed sphere activity by 55% (10 mm) and 40% (13 mm) in a standard tumor phantom.
- Tumor heterogeneity phantom results showed minimal accuracy improvement (<1%) after EM correction matrix changes fell below 35%.
- SV-PVC of patient tumors revealed frequent regional changes of +/-30% in heterogeneous areas.
Conclusions:
- The SV-PVC method, requiring only a tumor mask, automatically determines optimal iterations, unlike spatially invariant methods.
- SV-PVC improves quantitative accuracy in PET imaging of heterogeneous tumors, essential for reliable assessment and treatment planning.
- Correction for partial volume effects is recommended for research involving PET-imaged tumor heterogeneity.
