Attenuation Effects of MR Headphones During Brain PET/MR Studies

Aaron Ferguson1, Jonathan McConathy2, Yi Su2

  • 1Department of Medical Imaging and Radiation Therapeutics, Saint Louis University, St. Louis, Missouri; and laforestr@mir.wustl.edu aaroneferguson@yahoo.com.

Abstract

Insights

MR-safe headphones cause attenuation in PET/MR imaging, reducing activity concentration. A CT-derived map corrects this, minimizing the effect from 13.2% to 1.9% in phantom and human studies.

Area of Science:

  • Medical Imaging
  • Nuclear Medicine
  • Magnetic Resonance Imaging

Background:

  • Positron Emission Tomography/Magnetic Resonance (PET/MR) imaging offers advantages over PET/CT.
  • Attenuation correction in PET/MR is challenging due to the lack of direct correlation between MR signal and photon attenuation.
  • MR hardware, including headphones, can introduce significant attenuation artifacts.

Purpose of the Study:

  • To evaluate the impact of MR-compatible headphones on activity concentration measurements in PET/MR.
  • To develop and assess a practical method for correcting headphone-induced attenuation using CT-derived maps.

Main Methods:

  • Phantom studies using an (18)F-FDG filled phantom were conducted with and without MR-safe headphones.
  • Activity concentrations were compared between the two settings.
  • A CT-derived attenuation map of the headphones was created and applied to phantom and human brain data for correction.

Main Results:

  • Headphones reduced measured activity concentrations by up to 13.2% in phantom studies.
  • The proposed correction method reduced the maximum attenuation effect to 1.9%.
  • Similar attenuation effects and successful correction were observed in human brain imaging.

Conclusions:

  • MR-safe headphones are a notable source of attenuation in quantitative PET/MR brain imaging.
  • A CT-based attenuation correction approach effectively mitigates headphone-induced artifacts.
  • Accurate attenuation correction is crucial for reliable quantitative PET/MR studies.