Zhu, Jiening, Zhang, Chengzhu, Fan, Jason, Kuo, LiCheng, Cai, Weixing, He, Xiuxiu, Cervino, Laura, Moran, Jean, Li, Xiang, Li, Tianfang, Fu, Yabo · Oct 27, 2026
Jiening Zhu and coauthors present EUReCA for clinical CBCT reconstruction
Markdown · JSON · Original source
The paper’s authors present EUReCA, a model for reconstructing CBCT images from varying numbers of projections in clinical linac acquisitions. They say it uses one set of weights for 8–100 projections, without view-count-specific retraining, and can extend unchanged to other clinical trajectories. EUReCA combines learned projection features with an in-frame Feldkamp-Davis-Kress (FDK) reconstruction, then refines the reconstructed volume. The authors report that, on simulated data, it outperformed applicable feed-forward baselines and ran orders of magnitude faster than iterative optimization-based methods. In tests against vendor reconstructions of 15 measured patient CBCT scans, they report the best performance at every tested view count. The authors argue these results point to a practical, flexible approach to variable-view linac CBCT. The evidence described here is from simulated data and comparisons with vendor reconstructions; it does not establish how the model performs in clinical use. What performance would it achieve in clinical use beyond comparisons with vendor reconstructions?
Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.
Original source
cs.AI · Article
Source details
Original source date: Oct 27, 2026 (2026-10-27).
Date basis: Date basis not provided.
Added to Escape Meridian: 2026-11-11T12:55:33.926Z. First observed: 2026-10-30T06:43:03.346Z. Retrieved: 2026-10-31T05:31:16.313Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
The reported patient-scan comparison covers 15 measured scans evaluated against vendor reconstructions; the abstract does not report clinical-use outcomes. The abstract reports results on simulated data and comparisons with vendor reconstructions, but does not specify the tested view counts for the patient scans.
Open question
What performance would EUReCA achieve in clinical use beyond comparisons with vendor reconstructions?
Source-stated
The authors describe EUReCA as using one set of weights for 8–100 projections without view-count-specific retraining, and say its architecture and training recipe extend unchanged to other clinical trajectories.
For a given acquisition trajectory, EUReCA accepts 8-100 projections with a single set of weights, without view-count-specific retraining; the same architecture and training recipe extend unchanged to other clinical trajectories.
Source-stated
The authors report that EUReCA outperformed applicable feed-forward baselines on simulated data and ran orders of magnitude faster than iterative optimization-based methods.
On simulated data, EUReCA achieved 29.9-33.2 dB PSNR with 10-100 views and outperformed all applicable feed-forward baselines; it also exceeded iterative optimization-based methods while running orders of magnitude faster.
Source-stated
In comparisons against vendor reconstructions of 15 measured patient CBCT scans, the authors report that EUReCA had the best performance at every tested view count.
On 15 measured patient CBCT scans evaluated against vendor reconstructions, EUReCA achieved the best performance at every tested view count.
Source-stated
The paper presents geometry-aware, measurement-anchored feed-forward reconstruction as a possible foundation for variable-view linac CBCT.
These results suggest that geometry-aware, measurement-anchored feed-forward reconstruction can provide a practical and flexible foundation for variable-view linac CBCT.
Source and contributor context
Zhu, Jiening, Zhang, Chengzhu, Fan, Jason, Kuo, LiCheng, Cai, Weixing, He, Xiuxiu, Cervino, Laura, Moran, Jean, Li, Xiang, Li, Tianfang, Fu, Yabo · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
No suitable media was found by the supported page extraction.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
arXiv
arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Editorial provenance
Version 1. Publication ID: c8213d83-3b1e-4588-8279-59a6bbf2980a.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-11T12:55:33.926Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Observation: af766c2a4d2d0e6a7f30dd25d3938842c025b7c5321b32f2d9fd7cf09043c14b. Source text hash: d78f42a32468a9c90ead3bdade119f440fb1ba202cfa7a988068f2910d01dcfb.
Revision and source-check history
Version 1 · 2026-11-11T12:55:33.926Z
Jiening Zhu and coauthors present EUReCA for clinical CBCT reconstruction
The paper’s authors present EUReCA, a model for reconstructing CBCT images from varying numbers of projections in clinical linac acquisitions. They say it uses one set of weights for 8–100 projections, without view-count-specific retraining, and can extend unchanged to other clinical trajectories. EUReCA combines learned projection features with an in-frame Feldkamp-Davis-Kress (FDK) reconstruction, then refines the reconstructed volume. The authors report that, on simulated data, it outperformed applicable feed-forward baselines and ran orders of magnitude faster than iterative optimization-based methods. In tests against vendor reconstructions of 15 measured patient CBCT scans, they report the best performance at every tested view count. The authors argue these results point to a practical, flexible approach to variable-view linac CBCT. The evidence described here is from simulated data and comparisons with vendor reconstructions; it does not establish how the model performs in clinical use. What performance would it achieve in clinical use beyond comparisons with vendor reconstructions?
Original source
cs.AI · Article
Source details
Original source date: Oct 27, 2026 (2026-10-27).
Date basis: Date basis not provided.
Added to Escape Meridian: 2026-11-11T12:55:33.926Z. First observed: 2026-10-30T06:43:03.346Z. Retrieved: 2026-10-31T05:31:16.313Z.
Limits
The reported patient-scan comparison covers 15 measured scans evaluated against vendor reconstructions; the abstract does not report clinical-use outcomes. The abstract reports results on simulated data and comparisons with vendor reconstructions, but does not specify the tested view counts for the patient scans.
Open question
What performance would EUReCA achieve in clinical use beyond comparisons with vendor reconstructions?
Source-stated
The authors describe EUReCA as using one set of weights for 8–100 projections without view-count-specific retraining, and say its architecture and training recipe extend unchanged to other clinical trajectories.
For a given acquisition trajectory, EUReCA accepts 8-100 projections with a single set of weights, without view-count-specific retraining; the same architecture and training recipe extend unchanged to other clinical trajectories.
Source-stated
The authors report that EUReCA outperformed applicable feed-forward baselines on simulated data and ran orders of magnitude faster than iterative optimization-based methods.
On simulated data, EUReCA achieved 29.9-33.2 dB PSNR with 10-100 views and outperformed all applicable feed-forward baselines; it also exceeded iterative optimization-based methods while running orders of magnitude faster.
Source-stated
In comparisons against vendor reconstructions of 15 measured patient CBCT scans, the authors report that EUReCA had the best performance at every tested view count.
On 15 measured patient CBCT scans evaluated against vendor reconstructions, EUReCA achieved the best performance at every tested view count.
Source-stated
The paper presents geometry-aware, measurement-anchored feed-forward reconstruction as a possible foundation for variable-view linac CBCT.
These results suggest that geometry-aware, measurement-anchored feed-forward reconstruction can provide a practical and flexible foundation for variable-view linac CBCT.
Source and contributor context
Zhu, Jiening, Zhang, Chengzhu, Fan, Jason, Kuo, LiCheng, Cai, Weixing, He, Xiuxiu, Cervino, Laura, Moran, Jean, Li, Xiang, Li, Tianfang, Fu, Yabo · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
No suitable media was found by the supported page extraction.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
arXiv
arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Editorial provenance
Version 1. Publication ID: c8213d83-3b1e-4588-8279-59a6bbf2980a.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-11T12:55:33.926Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Observation: af766c2a4d2d0e6a7f30dd25d3938842c025b7c5321b32f2d9fd7cf09043c14b. Source text hash: d78f42a32468a9c90ead3bdade119f440fb1ba202cfa7a988068f2910d01dcfb.
uncertain · 2026-11-11T12:55:33.926Z · anthropic/claude-opus-5.5
Publication: c8213d83-3b1e-4588-8279-59a6bbf2980a
attribution — pass: The title and summary attribute EUReCA and its results to Jiening Zhu and coauthors. Performance claims and the closing argument are framed as the authors' reports and argument, not as established facts.
Authors:Jiening Zhu, Chengzhu Zhang, Jason Fan, LiCheng Kuo, Weixing Cai, Xiuxiu He, Laura Cervino, Jean Moran, Xiang Li, Tianfang Li, Yabo Fu
summary — uncertain: The reading never expands or explains CBCT (cone-beam computed tomography). It omits the clinical setting, image-guided radiotherapy, and the source's core stake: using fewer projections lowers dose and scan time, which makes reconstruction harder. Without this, a general reader cannot see why handling 8-100 projections matters. Linac, projections, FDK and feed-forward also go unexplained; these need term-library tooltips. Smallest repair: in the opening paragraph, expand CBCT and add the abstract's dose and scan-time motivation, attributed to the authors.
Cone-beam computed tomography (CBCT) is widely used for image-guided radiotherapy, but reducing the number of projections to lower dose and scan time makes reconstruction increasingly ill-posed.
date — pass: The originalPublished date of 2026-10-27 matches the arXiv v1 submission date. It is not the later feed date of 2026-10-31.
[Submitted on 27 Oct 2026]
support — pass: Each claim matches the abstract and is framed as source-stated. The limitations are accurate: the abstract gives no clinical-outcome data and does not list the view counts tested on patient scans. The caution that the results do not establish clinical-use performance follows from the evidence and does not invent any claim.
On 15 measured patient CBCT scans evaluated against vendor reconstructions, EUReCA achieved the best performance at every tested view count.
media — pass: The selected media is the verified direct arXiv abstract page for this paper. It is the most direct way for readers to access the source.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Explore arXivarXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Explore arXiv