Motivation: MRI-guided cardiac catheterization relies on manual slice tracking during catheter navigation, complicating procedures and limiting catheter visibility. An automated slice tracking approach has been proposed for continuous catheter visualization but relies on operator/subject-specific inputs. Goal(s): To develop and evaluate a real-time, parameter-free AI-based framework for continuous, automatic tracking of gadolinium-filled balloon catheters during MRI-guided cardiac catheterization. Approach: A deep-learning framework with two imaging modes was tested on a 3D-printed phantom and pediatric patients, assessing accuracy, sensitivity, and specificity. Results: The framework demonstrated high accuracy (>98%) and robustness across various anatomies and contrasts, showing feasibility for patient-independent tracking in MRI-guided cardiac procedures. Impact: This AI-based framework could simplify MRI-guided cardiac catheterization by reducing manual input, potentially decreasing procedural time and enhancing visualization accuracy, supporting broader clinical adoption and improved patient outcomes.
Neofytou et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: