Dynamic Angle Selection in X-Ray CT: A Reinforcement Learning Approach to Optimal Stopping

2025-03-16

AI Summary

To address the inherent accuracy-efficiency trade-off in sparse-angle scanning for industrial X-ray computed tomography (CT) online inspection, this paper proposes a reinforcement learning–based framework for dynamic angle selection and adaptive stopping. Methodologically, we are the first to embed optimal stopping theory into the sequential optimal experimental design (OED) framework, formulating an Actor-Critic differentiable policy gradient algorithm that jointly optimizes the projection angle sequence and termination timing. A differentiable CT reconstruction model is further introduced to enable end-to-end sim-to-real transfer. Evaluated on real industrial CT data, our approach significantly improves reconstruction quality and robustness—particularly under complex geometries and high noise—while enabling resource-aware, real-time scanning decisions. The framework thus achieves a principled balance between detection accuracy and operational efficiency.

Technology Category

• Search and Optimization: Learning to Search
• Reasoning under Uncertainty: Stochastic Optimization
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

In industrial X-ray Computed Tomography (CT), the need for rapid in-line inspection is critical. Sparse-angle tomography plays a significant role in this by reducing the required number of projections, thereby accelerating processing and conserving resources. Most existing methods aim to balance reconstruction quality and scanning time, typically relying on fixed scan durations. Adaptive adjustment of the number of angles is essential; for instance, more angles may be required for objects with complex geometries or noisier projections. The concept of optimal stopping, which dynamically adjusts this balance according to varying industrial needs, remains underutilized. Building on our previous work, we integrate optimal stopping into sequential Optimal Experimental Design (OED). We propose a novel method for computing the policy gradient within the Actor-Critic framework, enabling the development of adaptive policies for informative angle selection and scan termination. Additionally, we investigated the gap between simulation and real-world applications in the context of the developed learning-based method. Our trained model, developed using synthetic data, demonstrates reliable performance when applied to real-world data. This approach enhances the flexibility of CT operations and expands the applicability of sparse-angle tomography in industrial settings.

Problem

• Dynamic angle selection in X-ray CT for rapid inspection
• Balancing reconstruction quality and scanning time adaptively
• Bridging simulation-real-world gap in learning-based CT methods

Innovation

• Reinforcement learning optimizes X-ray CT angle selection.
• Actor-Critic framework enables adaptive scan termination policies.
• Synthetic data training ensures real-world CT application reliability.