An MR-Based Model for Cardio-Respiratory Motion Compensation of Overlays in X-Ray Fluoroscopy

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Radiology

Background:

• Static overlays in X-ray fluoroscopy lack accuracy due to patient motion.
• Cardiac and respiratory movements significantly degrade the precision of soft tissue visualization.

Purpose of the Study:

• To develop and evaluate a system for motion compensation of static overlays in X-ray fluoroscopy.
• To improve the accuracy of soft tissue visualization by accounting for cardiac and respiratory motion.

Main Methods:

• A 3-D motion model was created using real-time magnetic resonance (MR) imaging, with sagittal slices retrospectively stacked.
• Cardiac (ECG) and respiratory information from images were used for slice stacking and temporal smoothness enhancement.
• Deformable 3-D/3-D registration estimated motion from MR volumes; a linear direct correspondence model was applied in X-ray fluoroscopy using surrogate signals.

Main Results:

• The system reduced the 2-D Euclidean distance of annotated points from 3.85 to 2.75 mm in MR images.
• Accuracy in contrast-enhanced X-ray images improved, with point distance reduced from 3.0 to 1.8 mm.
• Qualitative evaluation demonstrated the effectiveness of motion-compensated overlays through images and videos.

Conclusions:

• The proposed system effectively compensates for cardiac and respiratory motion in X-ray fluoroscopy overlays.
• This motion compensation significantly enhances the accuracy and quality of soft tissue visualization.
• The real-time application using surrogate signals makes the system practical for clinical use.