Improving Generalizability in Limited-Angle CT Reconstruction with Sinogram Extrapolation
This addresses the problem of poor cross-dataset generalization in deep learning for CT reconstruction, which is incremental as it builds on existing methods with a new module.
The paper tackled limited-angle CT reconstruction by proposing a sinogram extrapolation module to improve model generalizability, achieving state-of-the-art performance on the NIH-AAPM dataset and significantly better generalization on unseen datasets like COVID-19 and LIDC.
Computed tomography (CT) reconstruction from X-ray projections acquired within a limited angle range is challenging, especially when the angle range is extremely small. Both analytical and iterative models need more projections for effective modeling. Deep learning methods have gained prevalence due to their excellent reconstruction performances, but such success is mainly limited within the same dataset and does not generalize across datasets with different distributions. Hereby we propose ExtraPolationNetwork for limited-angle CT reconstruction via the introduction of a sinogram extrapolation module, which is theoretically justified. The module complements extra sinogram information and boots model generalizability. Extensive experimental results show that our reconstruction model achieves state-of-the-art performance on NIH-AAPM dataset, similar to existing approaches. More importantly, we show that using such a sinogram extrapolation module significantly improves the generalization capability of the model on unseen datasets (e.g., COVID-19 and LIDC datasets) when compared to existing approaches.