OA-Recon: An Online Adaptive Reconstruction Framework for Sparse-View Multi-Energy CT
Abstract
Multi-energy CT (MECT) offers unique advantages in material decomposition, tissue characterization, and functional imaging, positioning it as a pivotal direction for next-generation CT. Currently, standardized scanning protocols for MECT have not yet been established. Considering growing public concern over X-ray radiation exposure, we propose a complementary sparse-view scanning protocol tailored for MECT, which reduces radiation dose while maximizing angular coverage. To reconstruct high-quality images from these sparse-view data and ensure algorithmic reliability in practical applications, we introduce an Online Adaptive Reconstruction (OA-Recon) framework that adapts robustly to varying acquisition settings through two designs. First, it adopts a Bayesian adaptation strategy for instance-specific optimization while preserving the learned prior. Second, it incorporates a Frequency-adaptive and Physics-informed Network (FaPiNet) for adaptive feature extraction and acquisition-conditioned feature modulation. In addition, it incorporates a spectral attention mechanism to fully exploit complementary information across energy channels. Experiments on simulated MECT and real mouse PCCT data show that FaPiNet-OA-Recon achieves better performance in suppressing streak artifacts, restoring image details, and maintaining CT-value accuracy. More importantly, OA-Recon demonstrates adaptability to changes in view, spectrum, and anatomy, providing a preliminarily feasible solution for clinical applications of MECT.
Related
- A UAV-based sparse-view 3DGS framework for greenhouse strawberry reconstruction
- Top2All: Benchmarking sparse-to-transcriptome reconstruction from high-expression profiles across sex- and sex-steroid-defined conditions
- Reconstruction of FACS-partitioned Adaptive Immune Receptor Repertoires from FACS-partitioned B and T Cell Subsets
- VizIt: A multi-view framework for exploring single-cell, spatial, and genetic data online
- Petimot: A Novel Framework for Inferring Protein Motions from Sparse Data Using SE(3)-Equivariant Graph Neural Networks
BibTeX
@article{wang2026oarecon,
title = {{OA-Recon: An Online Adaptive Reconstruction Framework for Sparse-View Multi-Energy CT}},
author = {Xi Wang and Tong Lin and Jiashun Wang and Xu Ji and Chun Yang and Tianyu Tang and Yikun Zhang and Jean-Louis Coatrieux and Jiasong Wu and Shenghong Ju and Yang Chen},
journal = {IEEE Transactions on Medical Imaging},
year = {2026},
month = oct,
doi = {10.1109/tmi.2026.3728200},
url = {https://hal.science/hal-05826365v1},
note = {CC BY-NC 4.0},
}
RIS
TY - JOUR
TI - OA-Recon: An Online Adaptive Reconstruction Framework for Sparse-View Multi-Energy CT
AU - Xi Wang
AU - Tong Lin
AU - Jiashun Wang
AU - Xu Ji
AU - Chun Yang
AU - Tianyu Tang
AU - Yikun Zhang
AU - Jean-Louis Coatrieux
AU - Jiasong Wu
AU - Shenghong Ju
AU - Yang Chen
PY - 2026
DA - 2026/10/23
JO - IEEE Transactions on Medical Imaging
DO - 10.1109/tmi.2026.3728200
UR - https://hal.science/hal-05826365v1
AB - Multi-energy CT (MECT) offers unique advantages in material decomposition, tissue characterization, and functional imaging, positioning it as a pivotal direction for next-generation CT. Currently, standardized scanning protocols for MECT have not yet been established. Considering growing public concern over X-ray radiation exposure, we propose a complementary sparse-view scanning protocol tailored for MECT, which reduces radiation dose while maximizing angular coverage. To reconstruct high-quality images from these sparse-view data and ensure algorithmic reliability in practical applications, we introduce an Online Adaptive Reconstruction (OA-Recon) framework that adapts robustly to varying acquisition settings through two designs. First, it adopts a Bayesian adaptation strategy for instance-specific optimization while preserving the learned prior. Second, it incorporates a Frequency-adaptive and Physics-informed Network (FaPiNet) for adaptive feature extraction and acquisition-conditioned feature modulation. In addition, it incorporates a spectral attention mechanism to fully exploit complementary information across energy channels. Experiments on simulated MECT and real mouse PCCT data show that FaPiNet-OA-Recon achieves better performance in suppressing streak artifacts, restoring image details, and maintaining CT-value accuracy. More importantly, OA-Recon demonstrates adaptability to changes in view, spectrum, and anatomy, providing a preliminarily feasible solution for clinical applications of MECT.
SN - 0278-0062
ER -