Related Experiment Video
Updated: Jun 12, 2026
A Randomized, Sham-Controlled Trial of Cranial Electrical Stimulation for Fibromyalgia Pain and Physical Function, Using Brain Imaging Biomarkers
Published on: January 5, 2024
Optimized EPI for fMRI using a slice-dependent template-based gradient compensation method to recover local
Jochen Rick1, Oliver Speck, Simon Maier
1Department of Radiology, Medical Physics, University Hospital Freiburg, Freiburg, Germany. jochen.rick@uniklinik-freiburg.de
Object:
Most functional magnetic resonance imaging (fMRI) experiments use gradient-echo echo planar imaging (GE EPI) to detect the blood oxygenation level-dependent (BOLD) effect. This technique may fail in the presence of anatomy-related susceptibility-induced field gradients in the human head. In this work, we present a novel 3D compensation method in combination with a template-based correction that can be optimized over particular regions of interest to recover susceptibility-induced signal loss without acquisition time penalty.
Materials And Methods:
Based on an evaluation of B(0) field maps of eight subjects, slice-dependent gradient compensation moments are derived for maximal BOLD sensitivity in two compromised regions: the orbitofrontal cortex and the amygdala areas. A modified EPI sequence uses these additional gradient moments in all three imaging directions. The method is compared to non-compensated, template-based and subject-specific correction gradients and also in a breath-holding experiment.
Results:
The slice-dependent gradient compensation method significantly improves signal intensity/BOLD sensitivity by about 35/43% in the orbitofrontal cortex and by 17/30% in the amygdala areas compared to a conventional acquisition. Template-based correction and subject-specific correction perform equally well. The BOLD sensitivity in the breath hold experiment is effectively increased in compensated regions.
Conclusion:
The new method addresses the problem of susceptibility-induced signal loss, without compromising temporal resolution. It can be used for event-related functional experiments without requiring additional subject-specific calibration or calculation time.
