Simultaneous estimation of a model-derived input function for quantifying cerebral glucose metabolism with [¹⁸F]FDG...

Area of Science:

• Nuclear Medicine
• Neuroimaging
• Biophysics

Background:

• Dynamic [18F]FDG PET requires invasive arterial sampling for accurate cerebral metabolic rate of glucose (CMRGlu) quantification.
• Simultaneous estimation (SIME) models the input function from image data, but traditional methods assume a specific mathematical form.
• A novel SIME approach utilizes a two-tissue compartment model for a more robust input function estimation.

Purpose of the Study:

• To introduce and evaluate a model-derived input function (MDIF) approach for dynamic [18F]FDG PET.
• To assess the MDIF approach's accuracy and reliability in both animal and human studies.
• To explore the feasibility of generating parametric images using the MDIF.

Main Methods:

• Simulations were conducted to validate the MDIF approach's accuracy.
• Animal studies compared MDIFs to measured arterial input functions (AIFs).
• Human studies (n=18) compared the MDIF method to the standard SIME-IDIF using dynamic [18F]FDG PET data.

Main Results:

• Simulations confirmed accurate MDIF extraction from whole-brain time activity curves.
• Animal experiments showed good agreement between MDIFs and AIFs.
• Human data demonstrated comparable CMRGlu values between MDIF and IDIF methods, with MDIF better characterizing early tracer kinetics.

Conclusions:

• The proposed model-driven SIME method successfully derives high signal-to-noise ratio (SNR) input functions.
• The MDIF approach offers advantages over traditional SIME, requiring fewer parameters and accommodating diverse input function shapes.
• MDIF's high SNR facilitates voxelwise parameter extraction, especially when combined with advanced estimation techniques like variational Bayesian methods.