A novel population-based input function integral estimation method for K i $K_i$ parametric imaging with shortened...

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Radiochemistry

Background:

• Dynamic positron emission tomography (PET) is crucial for clinical tumor diagnosis.
• Conventional dynamic PET scans are lengthy (approx. 60 min), causing patient inconvenience and limiting widespread use.

Purpose of the Study:

• To develop an innovative method for shortened, high-quality kinetic parameter (Ki) imaging using the Patlak model.
• To reduce scan times for dynamic PET while maintaining diagnostic accuracy.

Main Methods:

• Proposed the population-based input function integral estimation (PBIF-IE) method.
• Developed a linear regression model correlating early-stage integral (S_early) and late-stage mean (M_late) of image-derived input functions (IDIFs).
• Validated the method using datasets with varied framing protocols and evaluated performance across 30, 20, and 10-minute scan durations using PSNR, SSIM, and RE metrics.

Main Results:

• PBIF-IE method demonstrated superior performance in S_early estimation and Ki parametric imaging compared to other methods.
• 30-minute scans yielded Ki images highly consistent with 60-minute scans.
• 20-minute scans proved sufficient for preliminary tumor localization.

Conclusions:

• The PBIF-IE method is effective for shortened Ki parametric imaging, outperforming existing techniques.
• Future research will focus on optimizing the model by analyzing the impact of training dataset size on performance.