Improved [⁶⁸Ga]Ga-PSMA-11 image qualities reconstructed by total variation regularized expectation maximization on...

Area of Science:

• Medical Imaging
• Radiopharmaceuticals
• Oncology

Background:

• Gallium (⁶⁸GA) prostate-specific membrane antigen-11 ([⁶⁸Ga]Ga-PSMA-11) is crucial for prostate cancer imaging.
• Total-body positron emission tomography/computed tomography (PET/CT) offers enhanced sensitivity and speed.
• Image reconstruction algorithms significantly impact diagnostic accuracy.

Purpose of the Study:

• To evaluate the image quality of the Total Variation regularized Expectation Maximization (TVREM) algorithm for [⁶⁸Ga]Ga-PSMA-11 total-body PET/CT.
• To compare TVREM performance against the conventional Ordered Subset Expectation Maximization (OSEM) algorithm.
• To determine optimal TVREM penalization factors for improved image reconstruction.

Main Methods:

• Phantom and patient [⁶⁸Ga]Ga-PSMA-11 total-body PET/CT datasets were reconstructed using TVREM with varying penalization factors (0.09-0.45) and acquisition times (20-300s).
• Comparison data were generated using OSEM (3 iterations, 20 subsets, 300s acquisition).
• Image quality metrics included contrast recovery (CR), tumor-to-background ratio (TBR), image noise, and maximum standardized uptake value (SUV_max). Clinical performance was assessed by radiologists.

Main Results:

• TVREM with 120s acquisition and penalization factors of 0.27-0.45 demonstrated superior CR, TBR, and SUV_max gain compared to OSEM (300s).
• Optimized TVREM (120s, penalization 0.36) achieved comparable image noise to OSEM (300s) while increasing SUV_max by 28% and reducing noise by 14%.
• Qualitative scores indicated TVREM (120s, penalization 0.36) was equivalent to OSEM (300s) and TVREM (300s).

Conclusions:

• The TVREM algorithm, with optimized penalization, can achieve high-quality [⁶⁸Ga]Ga-PSMA-11 total-body PET/CT images.
• TVREM enables comparable image quality to OSEM but with significantly reduced acquisition times.
• This suggests TVREM's potential for more efficient and effective prostate cancer imaging.