High-resolution motion compensation for brain PET imaging using real-time electromagnetic motion tracking.

Area of Science:

• Medical Imaging
• Neuroscience
• Biophysics

Background:

• High-resolution brain PET imaging is challenged by head motion, a primary source of image blur.
• Accurate real-time head position and orientation estimation are crucial for motion compensation in PET.

Purpose of the Study:

• Develop a high-resolution electromagnetic motion tracking (EMMT) system for PET-CT scanners.
• Implement event-by-event motion correction to enhance PET imaging.

Main Methods:

• EMMT system uses a transmitter (source) and sensor array to detect real-time head movements.
• Motion data is synchronized with PET list-mode data and transformed into scanner coordinates.
• Event-by-event rebinning corrects for rigid motion and depth-of-interaction (DOI) parallax.

Main Results:

• Prism-PET/EMMT system achieved tracking accuracy with RMS errors of 0.49° for rotation and 0.15 mm for translation.
• Motion correction reduced point source FWHM differences to 3.9% in Prism-PET and 14% in mCT/EMMT.
• EMMT reduced spill-in artifacts in [18F]MK6240 PET imaging, lowering lateral ventricle SUVR from 0.70 to 0.34.

Conclusions:

• The EMMT system offers a cost-effective, high frame-rate, non-line-of-sight solution for motion tracking.
• It achieves high rotational and translational tracking accuracy, effectively mitigating motion-induced blur.
• EMMT is suitable for high-resolution brain-dedicated PET scanners.