mlsanityData to 14 Oct 2026

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

arXiv 2506.02093 ↗First posted 2 Jun 2025eess.IV cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 30 of its 66 references that are in the map. First posted on arXiv 2 Jun 2025.

Builds on

30 of 66 references.

References · 30

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  4. 2021Image Super-Resolution via Iterative Refinementcited by 1,511
  5. 2018Demystifying MMD GANscited by 1,293
  6. 2022TensoRF: Tensorial Radiance Fieldscited by 1,280
  7. 2020clDice -- A Novel Topology-Preserving Loss Function for Tubular Structure Segmentationcited by 294
  8. 2024nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentationcited by 240
  9. 2023CLIP-Driven Universal Model for Organ Segmentation and Tumor Detectioncited by 224
  10. 2013Gradient Magnitude Similarity Deviation: A Highly Efficient Perceptual Image Quality Indexcited by 195
  11. 2023MedNeXt: Transformer-driven Scaling of ConvNets for Medical Image Segmentationcited by 162
  12. 2022NAF: Neural Attenuation Fields for Sparse-View CBCT Reconstructioncited by 108
18 more
  1. 2023STU-Net: Scalable and Transferable Medical Image Segmentation Models Empowered by Large-Scale Supervised Pre-trainingcited by 100
  2. 2023Structure-Aware Sparse-View X-ray 3D Reconstructioncited by 72
  3. 2024R2-Gaussian: Rectifying Radiative Gaussian Splatting for Tomographic Reconstructioncited by 67
  4. 2024Towards Generalizable Tumor Synthesiscited by 65
  5. 2024AbdomenAtlas: A Large-Scale, Detailed-Annotated, & Multi-Center Dataset for Efficient Transfer Learning and Open Algorithmic Benchmarkingcited by 65
  6. 2024Medformer: A Multi-Granularity Patching Transformer for Medical Time-Series Classificationcited by 53
  7. 2024Universal and Extensible Language-Vision Models for Organ Segmentation and Tumor Detection from Abdominal Computed Tomographycited by 31
  8. 2024Touchstone Benchmark: Are We on the Right Way for Evaluating AI Algorithms for Medical Segmentation?cited by 31
  9. 2024Stable Diffusion Segmentation for Biomedical Images with Single-step Reverse Processcited by 16
  10. 2024Text-Driven Tumor Synthesiscited by 13
  11. 2024DIFR3CT: Latent Diffusion for Probabilistic 3D CT Reconstruction from Few Planar X-Rayscited by 8
  12. 2024Global-Local Image Perceptual Score (GLIPS): Evaluating Photorealistic Quality of AI-Generated Imagescited by 7
  13. 2025X-LRM: X-ray Large Reconstruction Model for Extremely Sparse-View Computed Tomography Recovery in One Secondcited by 6
  14. 2024EAGLE: An Edge-Aware Gradient Localization Enhanced Loss for CT Image Reconstructioncited by 5
  15. 2024MSDiff: Multi-Scale Diffusion Model for Ultra-Sparse View CT Reconstructioncited by 4
  16. 2024Enhancing signal detectability in learning-based CT reconstruction with a model observer inspired loss functioncited by 2
  17. 2019Arbitrarily large iterative tomographic reconstruction on multiple GPUs using the TIGRE toolboxcited by 1
  18. 2024CT-SDM: A Sampling Diffusion Model for Sparse-View CT Reconstruction across All Sampling Ratescited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 1.

6 papers.

  1. 2025Scaling Tumor Segmentation: Best Lessons from Real and Synthetic Datacited by 13
  2. 2025MedGS: Gaussian Splatting for Multi-Modal 3D Medical Imagingcited by 3
  3. 2025See More, Change Less: Anatomy-Aware Diffusion for Contrast Enhancementcited by 3
  4. 2025Auditing Significance, Metric Choice, and Demographic Fairness in Medical AI Challengescited by 2
  5. 2025Inference-Time Search Using Side Information for Diffusion-Based Image Reconstructioncited by 1
  6. 2026Distilling Photon-Counting CT into Routine Chest CT through Clinically Validated Degradation Modelingcited by 0