Are Pixel-Wise Metrics Reliable for Sparse-View Computed Tomography Reconstruction?

2025-06-02

AI Summary

In sparse-view CT reconstruction, conventional pixel-level metrics (e.g., PSNR, SSIM) inadequately reflect anatomical integrity—particularly for thin, critical structures such as small organs, bowel loops, and vasculature. To address this, we propose CARE: an anatomy-aware, model-agnostic framework for completeness-aware reconstruction enhancement. CARE introduces the first organ-level anatomical integrity quantification metrics, explicitly distinguishing large organs, small organs, bowel, and vessels. It incorporates a differentiable structural integrity loss, synergizing segmentation-guided completeness measurement with multi-scale structural regularization. Crucially, CARE is compatible with analytical, implicit, and generative reconstruction models in a plug-and-play manner. Extensive experiments demonstrate substantial improvements in structural integrity across diverse reconstruction methods: +32% for large organs, +22% for small organs, +40% for bowel, and +36% for vessels—thereby overcoming the limitations of traditional pixel-fidelity–driven evaluation and optimization.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Widely adopted evaluation metrics for sparse-view CT reconstruction--such as Structural Similarity Index Measure and Peak Signal-to-Noise Ratio--prioritize pixel-wise fidelity but often fail to capture the completeness of critical anatomical structures, particularly small or thin regions that are easily missed. To address this limitation, we propose a suite of novel anatomy-aware evaluation metrics designed to assess structural completeness across anatomical structures, including large organs, small organs, intestines, and vessels. Building on these metrics, we introduce CARE, a Completeness-Aware Reconstruction Enhancement framework that incorporates structural penalties during training to encourage anatomical preservation of significant structures. CARE is model-agnostic and can be seamlessly integrated into analytical, implicit, and generative methods. When applied to these methods, CARE substantially improves structural completeness in CT reconstructions, achieving up to +32% improvement for large organs, +22% for small organs, +40% for intestines, and +36% for vessels.

Problem

• Assessing structural completeness in sparse-view CT reconstructions
• Proposing anatomy-aware metrics for evaluating critical anatomical structures
• Enhancing reconstruction quality via CARE framework for anatomical preservation

Innovation

• Proposed anatomy-aware evaluation metrics for CT
• Introduced CARE framework for structural preservation
• Model-agnostic CARE improves reconstruction completeness significantly