View in EDS

Few‐shot CBCT‐based synthetic CT generation with denoising diffusion probabilistic model.

Saved in:
Bibliographic Details
Title: Few‐shot CBCT‐based synthetic CT generation with denoising diffusion probabilistic model.
Authors: Yeap, Ping Lin1 (AUTHOR) ply21@cam.ac.uk, Du, Xin2 (AUTHOR), Zhou, Meng3 (AUTHOR), Hoole, Andrew4 (AUTHOR), Barnett, Gillian C.5 (AUTHOR), Jena, Rajesh1,5 (AUTHOR)
Source: Medical Physics. Nov2025, Vol. 52 Issue 11, p1-15. 15p.
Subjects: Cone beam computed tomography, Radiotherapy, Image quality analysis, Machine learning, Computed tomography, Tumor diagnosis, Data augmentation, Stochastic models
Abstract: Background: In adaptive cancer radiotherapy, treatment plans were periodically re‐optimised according to anatomical variations, with the goal of maintaining or improving clinical outcomes. This required high‐quality in‐room imaging for daily monitoring, tumour/organ segmentation and dose evaluation. However, in‐room cone‐beam computed tomography (CBCT) images suffered from inaccurate Hounsfield Unit (HU) values and poor image quality due to higher artifact presence and lower signal‐to‐noise ratio. While CBCT‐to‐CT generation models have shown promising results, many conventional networks require large datasets for training. Purpose: To address this, we proposed a denoising diffusion probabilistic model (DDPM) with a modified sampling process to enable few‐shot synthetic CT (sCT) generation from CBCT images. Methods: We used a denoising diffusion probabilistic model trained on rigidly registered CT‐CBCT image pairs in our approach, without requiring deformable registrations. Our model was trained on 25 patients from a head‐and‐neck cancer cohort at our institution, and validated and tested on separate sets of seven and eight patients. Our sampling method leveraged channel‐ and noised‐conditioning, demonstrating high‐quality sCT generation with limited training data. By adding noise to guiding CBCT images during sampling, we exploited the convergence of CBCT and planning CT (pCT) representations in latent space. This allowed sampling from a noisy representation of the data, achieving sCTs with high fidelity. Results: Evaluation on a head‐and‐neck (H&N) cancer dataset showed that sCTs outperformed CBCTs, reducing masked mean absolute error (MAE) from 131 ± 17 HU to 49 ± 7 HU, and improving peak signal‐to‐noise ratio (PSNR) from 20.0 ± 0.9 dB to 22.9 ± 1.1 dB and normalised cross‐correlation (NCC) from 0.93 ± 0.01 to 0.96 ± 0.01. We further evaluated the method's generalisability on phantoms and publicly available patient datasets. Without retraining, the model achieved moderate improvements over CBCT (e.g., masked MAE 87 ± 17 HU, PSNR 19.4 ± 1.7 dB, NCC 0.93 ± 0.04) on the external H&N dataset. However, after retraining on 27 in‐distribution patient cases, the model achieved comparable performance to our internal dataset, with MAE 48 ± 12 HU, PSNR 22.2 ± 2.5 dB, and NCC 0.95 ± 0.03. We also extended the model to the pelvis site, achieving similarly strong results (masked MAE 44 ± 9 HU, PSNR 21.9 ± 2.1 dB, NCC 0.96 ± 0.03), demonstrating the feasibility of our method across multiple anatomical sites. Visual inspections showed improved image quality, significantly reduced artifacts, and better anatomical preservation in sCTs. Clinician evaluations revealed a higher preference to use sCTs over pCTs for fractional evaluation. Conclusions: Our work offered a practical, data‐efficient and site‐robust solution for clinics to generate high‐fidelity sCTs, facilitating CBCT‐based dose evaluation and plan adaptation in adaptive radiotherapy workflows. [ABSTRACT FROM AUTHOR]
Copyright of Medical Physics is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 189333908
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Few‐shot CBCT‐based synthetic CT generation with denoising diffusion probabilistic model.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Yeap%2C+Ping+Lin%22">Yeap, Ping Lin</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> ply21@cam.ac.uk</i><br /><searchLink fieldCode="AR" term="%22Du%2C+Xin%22">Du, Xin</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhou%2C+Meng%22">Zhou, Meng</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Hoole%2C+Andrew%22">Hoole, Andrew</searchLink><relatesTo>4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Barnett%2C+Gillian+C%2E%22">Barnett, Gillian C.</searchLink><relatesTo>5</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Jena%2C+Rajesh%22">Jena, Rajesh</searchLink><relatesTo>1,5</relatesTo> (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Medical+Physics%22">Medical Physics</searchLink>. Nov2025, Vol. 52 Issue 11, p1-15. 15p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Cone+beam+computed+tomography%22">Cone beam computed tomography</searchLink><br /><searchLink fieldCode="DE" term="%22Radiotherapy%22">Radiotherapy</searchLink><br /><searchLink fieldCode="DE" term="%22Image+quality+analysis%22">Image quality analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br /><searchLink fieldCode="DE" term="%22Computed+tomography%22">Computed tomography</searchLink><br /><searchLink fieldCode="DE" term="%22Tumor+diagnosis%22">Tumor diagnosis</searchLink><br /><searchLink fieldCode="DE" term="%22Data+augmentation%22">Data augmentation</searchLink><br /><searchLink fieldCode="DE" term="%22Stochastic+models%22">Stochastic models</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Background: In adaptive cancer radiotherapy, treatment plans were periodically re‐optimised according to anatomical variations, with the goal of maintaining or improving clinical outcomes. This required high‐quality in‐room imaging for daily monitoring, tumour/organ segmentation and dose evaluation. However, in‐room cone‐beam computed tomography (CBCT) images suffered from inaccurate Hounsfield Unit (HU) values and poor image quality due to higher artifact presence and lower signal‐to‐noise ratio. While CBCT‐to‐CT generation models have shown promising results, many conventional networks require large datasets for training. Purpose: To address this, we proposed a denoising diffusion probabilistic model (DDPM) with a modified sampling process to enable few‐shot synthetic CT (sCT) generation from CBCT images. Methods: We used a denoising diffusion probabilistic model trained on rigidly registered CT‐CBCT image pairs in our approach, without requiring deformable registrations. Our model was trained on 25 patients from a head‐and‐neck cancer cohort at our institution, and validated and tested on separate sets of seven and eight patients. Our sampling method leveraged channel‐ and noised‐conditioning, demonstrating high‐quality sCT generation with limited training data. By adding noise to guiding CBCT images during sampling, we exploited the convergence of CBCT and planning CT (pCT) representations in latent space. This allowed sampling from a noisy representation of the data, achieving sCTs with high fidelity. Results: Evaluation on a head‐and‐neck (H&N) cancer dataset showed that sCTs outperformed CBCTs, reducing masked mean absolute error (MAE) from 131 ± 17 HU to 49 ± 7 HU, and improving peak signal‐to‐noise ratio (PSNR) from 20.0 ± 0.9 dB to 22.9 ± 1.1 dB and normalised cross‐correlation (NCC) from 0.93 ± 0.01 to 0.96 ± 0.01. We further evaluated the method's generalisability on phantoms and publicly available patient datasets. Without retraining, the model achieved moderate improvements over CBCT (e.g., masked MAE 87 ± 17 HU, PSNR 19.4 ± 1.7 dB, NCC 0.93 ± 0.04) on the external H&N dataset. However, after retraining on 27 in‐distribution patient cases, the model achieved comparable performance to our internal dataset, with MAE 48 ± 12 HU, PSNR 22.2 ± 2.5 dB, and NCC 0.95 ± 0.03. We also extended the model to the pelvis site, achieving similarly strong results (masked MAE 44 ± 9 HU, PSNR 21.9 ± 2.1 dB, NCC 0.96 ± 0.03), demonstrating the feasibility of our method across multiple anatomical sites. Visual inspections showed improved image quality, significantly reduced artifacts, and better anatomical preservation in sCTs. Clinician evaluations revealed a higher preference to use sCTs over pCTs for fractional evaluation. Conclusions: Our work offered a practical, data‐efficient and site‐robust solution for clinics to generate high‐fidelity sCTs, facilitating CBCT‐based dose evaluation and plan adaptation in adaptive radiotherapy workflows. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Medical Physics is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=189333908
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1002/mp.70126
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 1
    Subjects:
      – SubjectFull: Cone beam computed tomography
        Type: general
      – SubjectFull: Radiotherapy
        Type: general
      – SubjectFull: Image quality analysis
        Type: general
      – SubjectFull: Machine learning
        Type: general
      – SubjectFull: Computed tomography
        Type: general
      – SubjectFull: Tumor diagnosis
        Type: general
      – SubjectFull: Data augmentation
        Type: general
      – SubjectFull: Stochastic models
        Type: general
    Titles:
      – TitleFull: Few‐shot CBCT‐based synthetic CT generation with denoising diffusion probabilistic model.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Yeap, Ping Lin
      – PersonEntity:
          Name:
            NameFull: Du, Xin
      – PersonEntity:
          Name:
            NameFull: Zhou, Meng
      – PersonEntity:
          Name:
            NameFull: Hoole, Andrew
      – PersonEntity:
          Name:
            NameFull: Barnett, Gillian C.
      – PersonEntity:
          Name:
            NameFull: Jena, Rajesh
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 11
              Text: Nov2025
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 00942405
          Numbering:
            – Type: volume
              Value: 52
            – Type: issue
              Value: 11
          Titles:
            – TitleFull: Medical Physics
              Type: main
ResultId 1