A statistical reconstruction algorithm for positronium lifetime imaging using time-of-flight positron emission...

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Biophysics

Background:

• Positron emission tomography (PET) is crucial for diagnosing diseases like cancer and Alzheimer's.
• Positronium lifetime imaging (PLI) extends PET by probing tissue microenvironment properties.
• Accurate PLI reconstruction is challenging due to finite time-of-flight (TOF) resolution in PET systems.

Purpose of the Study:

• To develop an accurate method for reconstructing 2D positronium lifetime images from TOF PET data.
• To address the challenge of finite TOF resolution in PLI.
• To evaluate a novel maximum likelihood estimation (MLE) approach for PLI.

Main Methods:

• Introduced a maximum likelihood estimation (MLE) method for 2D PLI.
• Utilized an exponentially modified Gaussian (EMG) probability distribution to model positronium lifetime data from TOF PET.
• Compared the EMG-based MLE method with exponential likelihood and penalized surrogate methods.

Main Results:

• Computer simulations demonstrated that the proposed EMG-MLE method yields quantitatively accurate lifetime images.
• The EMG-MLE method effectively handles PLI data with finite TOF resolution.
• The method's capability to manage multiple positron populations was also shown.

Conclusions:

• The developed EMG-based MLE method offers a significant advancement for accurate positronium lifetime imaging.
• This technique enhances the diagnostic potential of PET by providing microenvironmental information.
• The method is robust and adaptable for complex positronium decay scenarios.