Are Pixel-Wise Metrics Reliable for Computerized Tomography Reconstruction?

Yuanhao Cai (Johns Hopkins University) · Alan Yuille (JHU) · Tianyu Lin (Johns Hopkins University) · Xinran Li (Yale University) · Chuntung Zhuang (Johns Hopkins University) · Qi Chen (Johns Hopkins University) · Kai Ding (Johns Hopkins University) · Zongwei Zhou (Johns Hopkins University)

analytical methodsanatomical preservationanatomical structuresanatomy-aware evaluation metricscompleteness-aware reconstruction enhancementct reconstructionsgenerative methodsimplicit methodsmodel-agnosticpeak signal-to-noise ratioperformance improvementsparse-view ct reconstructionstructural completenessstructural penaltiesstructural similarity index measure

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.