Toward on-the-fly trajectory optimization for C-arm CBCT under strong kinematic constraints

Area of Science:

• Medical Imaging
• Interventional Radiology
• Computational Imaging

Background:

• Cone beam computed tomography (CBCT) is crucial in interventional radiology for 3D imaging.
• Standard circular trajectories can be obstructed by patient size or equipment, leading to collisions.
• Pre-defined collision-free trajectories lack adaptability for unpredictable intra-procedural constraints.

Purpose of the Study:

• To develop an on-the-fly trajectory optimization strategy for CBCT.
• To enable collision-free imaging in scenarios where standard circular paths are not feasible.
• To maintain high image quality for the volume of interest (VOI) during dynamic adjustments.

Main Methods:

• Proposed a search strategy exploring collision-free areas and optimizing locally for image quality.
• Utilized simulations on a prior diagnostic CT scan as a digital phantom.
• Employed Feature SIMilarity Index (FSIM) as the primary objective function for trajectory selection.
• Evaluated performance using FSIM and Universal Quality Image (UQI) on anatomical targets within the Alderson-Rando phantom.

Main Results:

• Achieved comparable image quality in the VOI to standard C-arm circular CBCT.
• Demonstrated a relative deviation of less than 10% for FSIM and UQI metrics.
• The entire trajectory optimization process was completed in approximately three to four minutes.
• Successfully reconstructed images from optimized trajectories comparable to standard CBCT.

Conclusions:

• The proposed on-the-fly optimization strategy effectively generates collision-free CBCT trajectories.
• The method ensures high image quality comparable to standard techniques, even with unpredictable constraints.
• This approach offers a practical solution for challenging CBCT acquisitions in interventional radiology.