Structured Learning for Electromagnetic Field Modeling and Real-Time Inversion

TL;DR

This paper presents a neural network-based magnetic field modeling approach that overcomes the limitations of traditional models, enabling fast, accurate, and robust real-time magnetic field inversion for electromagnetic navigation systems.

Contribution

It introduces a multi-layer perceptron model that preserves linear current dependence, providing fast, accurate, and robust magnetic field inversion without strict physical assumptions.

Findings

Data-driven models match traditional model accuracy

Evaluation times are approximately 1 ms for real-time control

Design choices improve calibration stability

Abstract

Precise magnetic field modeling is fundamental to the closed-loop control of electromagnetic navigation systems (eMNS) and the analytical Multipole Expansion Model (MPEM) is the current standard. However, the MPEM relies on strict physical assumptions regarding source symmetry and isolation, and requires optimization-based calibration that is highly sensitive to initialization. These constraints limit its applicability to systems with complex or irregular coil geometries. This work introduces an alternative modeling paradigm based on multi-layer perceptrons that learns nonlinear magnetic mappings while strictly preserving the linear dependence on currents. As a result, the field models enable fast, closed-form minimum-norm inversion with evaluation times of approximately 1 ms, which is critical for high-bandwidth magnetic control. For model training and evaluation we use large-scale, high-density datasets collected from the research-grade OctoMag and clinical-grade Navion systems. Our results demonstrate that data-driven models achieve predictive fidelity equivalent to the MPEM while maintaining comparable data efficiency. Furthermore, we demonstrate that straightforward design choices effectively eliminate spurious workspace ill-conditioning frequently reported in MPEM-based calibration. To facilitate future research, we release the complete codebase and datasets open source.