Abstract 4366766: Computational Analysis of Right Ventricular Motion to Optimize Energy Harvesting in Leadless Pacemaker Design

Introduction: Leadless pacemakers (LP) have addressed many complications of traditional lead-based transvenous designs. However, their reliance on internal batteries remains a limitation. A piezoelectric approach—harnessing cardiac motion to power the device—has been proposed. Developing an effective energy-harvesting system requires quantitative insight into regional cardiac motion and how it translates into mechanical energy for power conversion. We chose to use cardiac MRI (cMRI) data to generate node-based right ventricle (RV) motion dynamics, enabling prediction of energy-harvesting potential and identification of optimal implant locations. Methods: Human heart cMRI data was obtained from the Human Heart Project ACDC database (n=6), while in-house pig cMRI data was used (n=3). The cMRI images were manually segmented using MeshLab (open-source mesh software), selected for its flexibility in node-based 3D analysis. The mesh at end-diastolic volume (EDV) was the sample mesh, with subsequent timepoints between EDV and end-systolic volume (ESV) as target meshes. For each node on the sample mesh, the shortest distance to each target mesh was computed to quantify nodal displacement. Although this approach primarily captures scalar displacement, it provides a tractable and consistent metric across subjects. Secondary motion parameters, including velocity and acceleration, were calculated using MATLAB. Results: We analyzed RV displacement along the interventricular septum— the primary site for LP implantation. The average maximum displacement in the upper, middle, and lower thirds of the septum was 2.0±2.7mm, 2.3±1.8mm, and 7.3±1.9mm, respectively. Displacement in the lower third was significantly higher than in the upper (p<0.01) and middle (p<0.05) thirds. Comparisons of human and pig cMRI data (with avg. subject weight of 64kg and 66kg, respectively) showed no significant differences in displacement or velocity, with only middle-third acceleration differing (p<0.05). Conclusion: This study introduces a computational pipeline for extracting detailed RV motion dynamics from cMRI. The key findings: 1) the lower third of the septum exhibits the greatest displacement, suggesting it may be the optimal site for maximizing energy harvesting in next-generation LP; and 2) similar motion dynamics in the region between humans and pigs support the pig as a valid preclinical model. This computational approach offers a foundational tool for energy-aware LP optimization.