Event-by-event respiratory motion correction for PET with 3D internal-1D external motion correlation

Area of Science:

• Nuclear Medicine
• Medical Imaging
• Image Reconstruction

Background:

• Respiratory motion during PET/CT imaging causes image blurring and underestimates tracer concentrations.
• Existing motion correction methods may not fully address both intraframe and interframe motion.
• Accurate quantification in static and dynamic PET/CT studies is crucial for diagnosis and treatment monitoring.

Purpose of the Study:

• To develop and evaluate an event-by-event respiratory motion correction method using three-dimensional internal-one-dimensional external motion correlation (INTEX3D).
• To fully correct for organ/tumor-specific rigid motion caused by respiration using all detected events.
• To investigate the quantitative improvement in static and dynamic PET/CT imaging.

Main Methods:

• Determined internal organ/tumor positional translation from phase-gated images using a level set segmentation method.
• Established linear correlations between 1D external respiratory signals and 3D internal organ centroids.
• Generated high-temporal-resolution 3D internal motion signals to guide event-by-event listmode reconstruction.

Main Results:

• INTEX3D increased contrast recovery by 5%-9% (vs. 1D correction) and 32%-40% (vs. no correction) in phantom studies.
• Mean organ activity increased by 9.3% (pancreas) and 11.2% (kidney) in human studies with [(18)F]FP(+)DTBZ.
• Average lesion-to-blood ratio increased by 20% in [(18)F]fluoromisonidazole (FMISO) studies; tracer concentration increased by 18% (pancreas) and 11% (kidney).

Conclusions:

• The INTEX3D method effectively compensates for both intra- and inter-gate respiratory motion while preserving all detected counts.
• The technique is applicable to both static and dynamic PET/CT imaging studies.
• Anatomical-based postfiltering further reduced noise by over 50% while maintaining quantitative accuracy and organ boundaries.