Rapid cardiac activation prediction for cardiac resynchronization therapy planning using geometric deep learning

2025-06-10

AI Summary

Approximately one-third of patients undergoing cardiac resynchronization therapy (CRT) exhibit nonresponse, largely due to suboptimal pacing site selection. Method: We propose a geometric deep learning framework for personalized ventricular activation time prediction and optimization. Specifically, we introduce the Geometric Information Neural Operator (GINO)—the first neural operator tailored for CRT planning—that enables mesh-invariant, end-to-end prediction of activation maps; it is trained on high-fidelity synthetic data generated via finite-element electrophysiological simulation. We further integrate an automated pacing site optimization pipeline with an interactive web-based GUI to support clinically interpretable decision-making. Contribution/Results: GINO achieves a mean absolute prediction error of 1.14%, substantially outperforming graph neural networks (3.14%). Optimized pacing reduces maximum activation time by 20%—versus only 8% under random selection. The system has been deployed as an online clinical decision support tool.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Learning to Search
• Constraint Satisfaction and Optimization: Constraint Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Cardiac resynchronization therapy (CRT) is a common intervention for patients with dyssynchronous heart failure, yet approximately one-third of recipients fail to respond due to suboptimal lead placement. Identifying optimal pacing sites remains challenging, largely due to patient-specific anatomical variability and the limitations of current individualized planning strategies. In a step towards constructing an in-silico approach to help address this issue, we develop two geometric deep learning (DL) models, based on graph neural network (GNN) and geometry-informed neural operator (GINO), to predict cardiac activation time map in real-time for CRT planning and optimization. Both models are trained on a large synthetic dataset generated from finite-element (FE) simulations over a wide range of left ventricular (LV) geometries, pacing site configurations, and tissue conductivities. The GINO model significantly outperforms the GNN model, with lower prediction errors (1.14% vs 3.14%) and superior robustness to noise and various mesh discretization. Using the GINO model, we also develop a workflow for optimizing the pacing site in CRT from given activation time map and LV geometry. Compared to randomly selecting a pacing site, the CRT optimization workflow produces a larger reduction in maximum activation time (20% vs. 8%). In conjunction with an interactive web-based graphical user interface (GUI) available at https://dcsim.egr.msu.edu/, the GINO model shows promising potential as a clinical decision-support tool for personalized pre-procedural CRT optimization.

Problem

• Predict cardiac activation time for CRT planning
• Optimize pacing site selection for CRT effectiveness
• Overcome patient-specific anatomical variability in CRT

Innovation

• Geometric deep learning predicts cardiac activation
• GINO model outperforms GNN with lower errors
• Interactive GUI for clinical CRT optimization