Development and validation of a variance model for dynamic PET: uses in fitting kinetic data and optimizing the...

Area of Science:

• Nuclear Medicine
• Medical Imaging
• Biophysics

Background:

• Dynamic Positron Emission Tomography (PET) data analysis relies on accurate biological parameter estimation.
• Estimate precision is often limited by the number of detected events, influenced by injected radiotracer activity (A(0)).

Purpose of the Study:

• To assess the benefits of optimizing injected activity (A(0)) for dynamic PET data.
• To evaluate a new variance model predicting estimate precision as a function of A(0).

Main Methods:

• Dynamic [(15)O]H(2)O PET scans were performed on seven cancer patients with varied A(0) (142-839 MBq).
• Data were analyzed using a new variance model and compared with simulations and patient data for accuracy.
• Parameter estimate precision was examined for different region sizes (ROIs).

Main Results:

• The new variance model accurately estimated relative variance in dynamic PET data across various A(0) and time frames.
• Patient data showed good agreement with model predictions for perfusion (F) and volume of distribution (V(T)) precision.
• Optimal A(0) of 500-700 MBq improved precision for small ROIs (<5 mL), reducing standard error.

Conclusions:

• Optimizing injected activity (A(0)) is crucial for enhancing biological parameter estimate precision in dynamic PET.
• The developed variance model effectively predicts precision improvements with optimized A(0).
• Optimal A(0) yields more accurate quantitative PET imaging, particularly for small target volumes.