Using Image Quality Metric to Improve Navigator-based Motion Correction Robustness

Motivation: Navigator-based rigid-body motion correction (MoCo) can improve IQ when the correct motion parameters are extracted from navigator data. However, navigator-based MoCo can reduce IQ when incorrect motion parameters are derived. Goal(s): We propose combining IQ assessment with navigator-based MoCo to improve correction robustness. Approach: The proposed solution was implemented by calculating the entropy (IQ metric) of images with/without navigator-based rigid-body MoCo, and MoCo was applied only when the image entropy decreased. Results: During volunteer studies, the proposed solution reduced the number of slices that showed IQ reduction due to rigid-body correction, while maintaining a large majority of slices that benefited from rigid-body correction. Impact: Navigator-based rigid-body motion correction (MoCo) robustness was improved by using an image quality metric to determine whether MoCo improved IQ.