Impact of attenuation correction on clinical [(18)F]FDG brain PET in combined PET/MRI

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Radiology

Background:

• Photon attenuation correction (AC) in PET/MRI is challenging due to the inability to directly derive linear coefficients.
• Cortical bone is typically excluded in current MR-AC methods, leading to potential underestimation of PET signals.
• Newer MR-AC techniques aim to incorporate bone information, but their clinical utility is under debate.

Purpose of the Study:

• To evaluate the quantitative accuracy and clinical impact of MR-AC in brain [18F]FDG PET/MRI compared to PET/CT.
• To determine if current quantitative errors due to bone omission in MR-AC affect diagnostic quality for dementia and movement disorders.

Main Methods:

• [18F]FDG PET/CT and PET/MRI scans were performed on 13 patients with suspected dementia or movement disorders.
• Multiple MR-AC methods were applied and compared against a CT-derived AC standard.
• Quantitative errors and diagnostic impact on hypometabolism patterns were assessed.

Main Results:

• Quantitative deviations from CT-AC ranged from -11.9% to +2%.
• These quantitative errors did not result in systematic diagnostic errors.
• Overall patterns of brain hypometabolism, crucial for clinical diagnosis, remained largely unchanged.

Conclusions:

• Clinical quality of brain [18F]FDG PET is not significantly affected by current MR-AC limitations, including bone omission.
• Brain [18F]FDG PET is suitable for clinical routine use even with suboptimal MR-AC.
• Improvements in MR-AC are still warranted for future applications.