Accelerated PET kinetic maps estimation by analytic fitting method.

Area of Science:

• Nuclear Medicine
• Pharmacokinetics
• Computational Modeling

Background:

• Dynamic Positron Emission Tomography (PET) generates complex time-activity curve (TAC) data.
• Accurate kinetic parameter estimation is crucial for quantitative PET imaging.
• Current methods for non-linear kinetic modeling can be computationally intensive.

Purpose of the Study:

• To develop and validate a novel, computationally efficient approach for non-linear kinetic parameter estimation from dynamic PET data.
• To compare the performance and computational speed of the proposed method against existing numerical algorithms.
• To assess the accuracy and clinical applicability of the new technique.

Main Methods:

• Derivation of an analytical closed-form expression for compartmental models using auxiliary parameters.
• Comparison with two alternative numerical calculation algorithms.
• Validation through simulation studies and analysis of dynamic 3D PET clinical data from control subjects.

Main Results:

• The proposed method achieved significant speed-up factors: approximately 120x compared to a fully numeric approach and 38x compared to a conventional implementation.
• Estimated model parameters using the new method closely matched those obtained by conventional techniques.
• Clinical data analysis corroborated simulation findings, demonstrating the method's viability.

Conclusions:

• The novel analytical approach offers a substantial reduction in computational burden for kinetic parameter estimation in dynamic PET.
• The method provides accurate parameter estimates comparable to existing techniques.
• This approach holds promise for improving the efficiency and applicability of quantitative PET in clinical practice.