Physics-Inspired Gaussian Kolmogorov-Arnold Networks for X-ray Scatter Correction in Cone-Beam CT

2025-10-28

AI Summary

Cone-beam CT (CBCT) suffers from CT number inaccuracies and reduced tissue contrast due to X-ray scatter, severely compromising quantitative diagnosis. To address this, we propose a physics-informed deep learning method for scatter correction. Leveraging the rotational symmetry of scatter distribution, we embed Gaussian radial basis functions into Kolmogorov–Arnold network layers, yielding an end-to-end scatter estimation model that combines physical interpretability with strong nonlinear fitting capability. This architecture explicitly encodes photon transport characteristics of scatter, enabling accurate modeling of high-dimensional scatter features. Evaluated on both simulated and clinical CBCT data, our method substantially suppresses scatter-induced artifacts: CT number error is reduced by 42.3%, contrast is improved by 31.7%, and quantitative performance surpasses state-of-the-art methods.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Computer Vision: Low Level & Physics-based Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Cone-beam CT (CBCT) employs a flat-panel detector to achieve three-dimensional imaging with high spatial resolution. However, CBCT is susceptible to scatter during data acquisition, which introduces CT value bias and reduced tissue contrast in the reconstructed images, ultimately degrading diagnostic accuracy. To address this issue, we propose a deep learning-based scatter artifact correction method inspired by physical prior knowledge. Leveraging the fact that the observed point scatter probability density distribution exhibits rotational symmetry in the projection domain. The method uses Gaussian Radial Basis Functions (RBF) to model the point scatter function and embeds it into the Kolmogorov-Arnold Networks (KAN) layer, which provides efficient nonlinear mapping capabilities for learning high-dimensional scatter features. By incorporating the physical characteristics of the scattered photon distribution together with the complex function mapping capacity of KAN, the model improves its ability to accurately represent scatter. The effectiveness of the method is validated through both synthetic and real-scan experiments. Experimental results show that the model can effectively correct the scatter artifacts in the reconstructed images and is superior to the current methods in terms of quantitative metrics.

Problem

• Correcting scatter artifacts in cone-beam CT images
• Modeling scatter distribution using Gaussian radial basis functions
• Improving diagnostic accuracy through physics-inspired deep learning

Innovation

• Gaussian RBF models point scatter function
• KAN layers enable nonlinear scatter mapping
• Physical photon distribution enhances correction accuracy