Maximum-Likelihood Joint Image Reconstruction/Motion Estimation in Attenuation-Corrected Respiratory Gated PET/CT...

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Image Reconstruction

Background:

• Positron emission tomography (PET) imaging is crucial for diagnosis and treatment monitoring.
• Motion artifacts, particularly from respiration, degrade PET image quality and quantification accuracy.
• Current motion correction methods often require complex, time-consuming, and potentially high-dose imaging protocols.

Purpose of the Study:

• To develop and evaluate a joint image reconstruction and motion estimation (JRM) method for PET.
• To investigate the feasibility of using a single, non-gated attenuation map (μ-map) for motion-corrected PET reconstruction.
• To compare the performance of JRM with traditional post-registration reconstruction and consolidation (PRRC) methods.

Main Methods:

• Utilized maximum likelihood (ML) principles for JRM, incorporating a probabilistic model that accounts for warped attenuation.
• Developed a joint optimization algorithm for JRM.
• Validated the algorithm using simulated respiratory-gated PET/CT data (XCAT phantom) and patient data.

Main Results:

• JRM demonstrated robustness, yielding similar reconstructed gates irrespective of the input μ-map's alignment.
• JRM significantly outperformed PRRC in hot lesion quantification in simulations.
• Reconstructed PET images using JRM with a breath-held μ-map closely matched those obtained with gated μ-maps.
• Patient data showed JRM with a breath-held high-resolution CT (HRCT) μ-map achieved comparable reconstruction quality to PRRC using cine-CT gated μ-maps.

Conclusions:

• A single breath-held μ-map is sufficient for accurate motion-corrected PET reconstruction using JRM.
• JRM offers a practical and potentially low-dose alternative to current motion correction techniques.
• This approach can enhance the implementation of motion-corrected respiratory-gated PET/CT imaging.