Pose Estimation for Intra-cardiac Echocardiography Catheter via AI-Based Anatomical Understanding

2025-05-07

HSMR 2025

AI Summary

Current ICE catheter navigation relies on electromagnetic tracking—prone to drift—or subjective manual adjustment, limiting clinical utility. This work proposes the first anatomy-aware, purely vision-based pose estimation method that enables real-time 3D localization and orientation estimation of the catheter within the left atrial coordinate system using only intracardiac echocardiography (ICE) images, without external sensors. Our approach leverages a Vision Transformer architecture enhanced with image patch embedding, dual-head [CLS] token regression, left atrial mesh normalization, and joint MSE optimization, trained on 851 clinically annotated cases. Quantitative evaluation yields mean localization error of 9.48 mm and orientation errors of 16.13°, 8.98°, and 10.47° across roll, pitch, and yaw, respectively; qualitative validation confirms high anatomical consistency between predicted and ground-truth views in 3D cardiac models. Hardware-agnostic, the method operates standalone or integrates seamlessly into platforms such as CARTO, advancing ICE navigation toward full automation and enhanced robustness.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Computer Vision: Biometrics, Face, Gesture & Pose
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Intra-cardiac Echocardiography (ICE) plays a crucial role in Electrophysiology (EP) and Structural Heart Disease (SHD) interventions by providing high-resolution, real-time imaging of cardiac structures. However, existing navigation methods rely on electromagnetic (EM) tracking, which is susceptible to interference and position drift, or require manual adjustments based on operator expertise. To overcome these limitations, we propose a novel anatomy-aware pose estimation system that determines the ICE catheter position and orientation solely from ICE images, eliminating the need for external tracking sensors. Our approach leverages a Vision Transformer (ViT)-based deep learning model, which captures spatial relationships between ICE images and anatomical structures. The model is trained on a clinically acquired dataset of 851 subjects, including ICE images paired with position and orientation labels normalized to the left atrium (LA) mesh. ICE images are patchified into 16x16 embeddings and processed through a transformer network, where a [CLS] token independently predicts position and orientation via separate linear layers. The model is optimized using a Mean Squared Error (MSE) loss function, balancing positional and orientational accuracy. Experimental results demonstrate an average positional error of 9.48 mm and orientation errors of (16.13 deg, 8.98 deg, 10.47 deg) across x, y, and z axes, confirming the model accuracy. Qualitative assessments further validate alignment between predicted and target views within 3D cardiac meshes. This AI-driven system enhances procedural efficiency, reduces operator workload, and enables real-time ICE catheter localization for tracking-free procedures. The proposed method can function independently or complement existing mapping systems like CARTO, offering a transformative approach to ICE-guided interventions.

Problem

• Eliminates need for external tracking sensors in ICE catheter navigation
• Overcomes interference and drift in electromagnetic tracking methods
• Provides real-time AI-based pose estimation from ICE images

Innovation

• AI-based pose estimation from ICE images
• Vision Transformer model for anatomical understanding
• Eliminates need for external tracking sensors