Physics-Informed Deep Contrast Source Inversion: A Unified Framework for Inverse Scattering Problems

2025-08-14

AI Summary

Inverse scattering problems in electromagnetic imaging and medical diagnostics face challenges including nonlinearity and diverse measurement conditions (e.g., full-field, phaseless, or multi-frequency data). To address these, this paper proposes DeepCSI—a physics-informed deep contrast source inversion framework. First, it integrates the contrast source inversion (CSI) method with neural operator principles, employing a ResMLP architecture to model excitation-dependent current distributions, thereby enabling effective linearization of nonlinear inverse problems. Second, it introduces learnable tensor representations for medium parameters, establishing an end-to-end differentiable framework that jointly optimizes network weights and physical parameters. Third, it designs a hybrid loss function incorporating the state equation, data fidelity term, and total-variation regularization. Experiments demonstrate that DeepCSI consistently outperforms conventional CSI methods across diverse measurement scenarios—achieving higher reconstruction accuracy, superior robustness, and significantly reduced computational cost.

Technology Category

• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Inverse scattering problems are critical in electromagnetic imaging and medical diagnostics but are challenged by their nonlinearity and diverse measurement scenarios. This paper proposes a physics-informed deep contrast source inversion framework (DeepCSI) for fast and accurate medium reconstruction across various measurement conditions. Inspired by contrast source inversion (CSI) and neural operator methods, a residual multilayer perceptron (ResMLP) is employed to model current distributions in the region of interest under different transmitter excitations, effectively linearizing the nonlinear inverse scattering problem and significantly reducing the computational cost of traditional full-waveform inversion. By modeling medium parameters as learnable tensors and utilizing a hybrid loss function that integrates state equation loss, data equation loss, and total variation regularization, DeepCSI establishes a fully differentiable framework for joint optimization of network parameters and medium properties. Compared with conventional methods, DeepCSI offers advantages in terms of simplicity and universal modeling capabilities for diverse measurement scenarios, including phase-less and multi-frequency observation. Simulations and experiments demonstrate that DeepCSI achieves high-precision, robust reconstruction under full-data, phaseless data, and multifrequency conditions, outperforming traditional CSI methods and providing an efficient and universal solution for complex inverse scattering problems.

Problem

• Solving nonlinear inverse scattering problems efficiently
• Reconstructing media accurately across diverse measurement scenarios
• Reducing computational cost of traditional full-waveform inversion

Innovation

• Physics-informed deep contrast source inversion framework
• ResMLP models current distributions, linearizes problem
• Hybrid loss function for joint optimization