Modelling approach for separating blood time-activity curves in positron emission tomographic studies

Area of Science:

• Nuclear Medicine
• Pharmacokinetics
• Biomedical Engineering

Background:

• Accurate quantification in Positron Emission Tomography (PET) relies on precise blood/plasma time-activity curves (TACs).
• Traditional methods for deriving these TACs can be complex and prone to errors, especially when dealing with metabolites.

Purpose of the Study:

• To develop and validate a novel modeling approach for generating complete radiotracer and metabolite time-activity curves from total radioactivity measurements.
• To improve the accuracy and reliability of quantitative PET studies.

Main Methods:

• A compartmental modeling technique was employed to simulate tracer-to-metabolite conversion within the body.
• The model utilizes the total radioactivity concentration curve as the input function.
• The approach was tested using two distinct PET imaging scenarios: 6-[18F]fluoro-L-dopa (FDOPA) kinetics and dynamic 15O oxygen PET.

Main Results:

• The modeling approach successfully derived the full plasma time-activity curves for FDOPA and its metabolites.
• In the 15O oxygen PET example, the method effectively solved a deconvolution problem, separating the time-activity curves of 15O oxygen and 15O water in blood.
• Demonstrated improved separation of blood/plasma TACs compared to conventional methods.

Conclusions:

• The developed modeling approach offers a robust method for obtaining accurate blood/plasma time-activity curves in PET studies.
• This technique enhances the quantitative interpretation of PET data, leading to more reliable research outcomes.