Device-Less Data-Driven Cardiac and Respiratory Gating Using LAFOV PET Histo Images

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Biophysics

Background:

• Modern Positron Emission Tomography (PET) excels at detecting small tumors but suffers from motion artifacts.
• Respiratory and cardiac motion degrade PET image quality, impacting diagnostic accuracy.
• Long Axial Field-of-View (LAFOV) PET scanners and precise Time-of-Flight (TOF) enable ultrafast imaging for motion correction.

Purpose of the Study:

• To develop and validate a device-less, data-driven method for estimating and correcting respiratory and cardiac motion in PET imaging.
• To assess the impact of motion correction on tumor visualization and quantitative metrics.

Main Methods:

• Generated ultrafast [18F]FDG PET histo-image series (0.25s frames).
• Estimated cardiac and respiratory frequencies using Short Time Fourier Transform (STFT) on 18 patients.
• Applied estimated frequencies for device-less, data-driven gated image reconstruction.

Main Results:

• Cardiac frequency estimates closely matched pulse oximeter references (p=0.97, 0.4 ± 0.3 bpm difference).
• Respiratory frequencies were within the expected range (10-20 rpm) in 16/18 patients.
• Gated reconstruction in three lung tumor patients showed increased SUVmax and decreased tumor volume compared to non-gated images.

Conclusions:

• The developed method successfully provides signals for gated PET reconstruction, correcting for cardiac and respiratory motion.
• This approach holds potential for significantly increasing diagnostic accuracy in PET imaging.
• Device-less motion estimation and correction can improve visualization and quantification of tumor lesions.