Prompting Lipschitz-constrained network for multiple-in-one sparse-view CT reconstruction

2025-11-25

AI Summary

In sparse-view CT reconstruction, deep unrolling methods face two key challenges: (i) the Lipschitz continuity of prior networks is difficult to verify theoretically, undermining convergence guarantees; and (ii) handling multiple sampling configurations requires separate models, incurring substantial memory overhead. To address these, we propose LipNet—the first prior network with *explicit, analytically verifiable* Lipschitz continuity and boundedness. We further design a learnable prompt module that injects sampling configuration information into the iterative reconstruction process, enabling a single model to adapt to diverse sparse-view settings. Built upon this, PromptCT ensures theoretical convergence while drastically reducing storage requirements. Experiments on both simulated and real-world data demonstrate superior reconstruction quality and strong generalization over state-of-the-art baselines. The code and datasets are publicly available.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Machine Learning: Calibration & Uncertainty Quantification
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Responsible Web: Measurement, analysis, and circumvention of Web censorship
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Despite significant advancements in deep learning-based sparse-view computed tomography (SVCT) reconstruction algorithms, these methods still encounter two primary limitations: (i) It is challenging to explicitly prove that the prior networks of deep unfolding algorithms satisfy Lipschitz constraints due to their empirically designed nature. (ii) The substantial storage costs of training a separate model for each setting in the case of multiple views hinder practical clinical applications. To address these issues, we elaborate an explicitly provable Lipschitz-constrained network, dubbed LipNet, and integrate an explicit prompt module to provide discriminative knowledge of different sparse sampling settings, enabling the treatment of multiple sparse view configurations within a single model. Furthermore, we develop a storage-saving deep unfolding framework for multiple-in-one SVCT reconstruction, termed PromptCT, which embeds LipNet as its prior network to ensure the convergence of its corresponding iterative algorithm. In simulated and real data experiments, PromptCT outperforms benchmark reconstruction algorithms in multiple-in-one SVCT reconstruction, achieving higher-quality reconstructions with lower storage costs. On the theoretical side, we explicitly demonstrate that LipNet satisfies boundary property, further proving its Lipschitz continuity and subsequently analyzing the convergence of the proposed iterative algorithms. The data and code are publicly available at https://github.com/shibaoshun/PromptCT.

Problem

• Ensuring Lipschitz constraints in deep unfolding networks for CT reconstruction
• Reducing storage costs by handling multiple sparse-view settings with one model
• Providing theoretical guarantees for network convergence and reconstruction quality

Innovation

• Lipschitz-constrained network ensures provable convergence
• Prompt module handles multiple sparse view configurations
• Single unified model reduces storage costs significantly