Improved quantitation for PET/CT image reconstruction with system modeling and anatomical priors

Area of Science:

• Medical Imaging
• Nuclear Medicine

Background:

• Accurate quantification of positron emission tomography (PET) tracer uptake is vital for cancer staging and treatment response monitoring.
• Quantitative accuracy in PET is significantly limited for small tumors due to partial volume effects and image smoothing.

Purpose of the Study:

• To develop a novel PET reconstruction algorithm to minimize quantitative errors caused by limited system resolution and noise.
• To improve the accuracy of tumor uptake measurements in PET imaging.

Main Methods:

• Incorporation of detection system response into the projection matrix of a statistical reconstruction algorithm.
• Utilizing aligned anatomical information from PET/CT scanners to guide image noise reduction via the penalty term.
• Combining these enhancements with Fourier rebinning for a clinically feasible, fully 3D PET reconstruction algorithm.

Main Results:

• The proposed method demonstrated improved quantitation of tumor values, reducing both bias and variance across various tumor sizes and activity levels.
• Compared to filtered back-projection, the algorithm reduced errors in maximum tumor values by 11% at typical clinical noise levels.
• A 5% reduction in error was observed compared to conventional iterative reconstruction methods.

Conclusions:

• The developed PET reconstruction algorithm effectively reduces quantitative errors in tumor uptake measurements.
• This method offers improved accuracy for small tumor imaging, benefiting clinical applications in cancer diagnosis and therapy monitoring.