EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Quantifying the impact of tumor size and motion on 4DCT‐4DCBCT image registration accuracy using machine learning and statistical analysis

Qiaoyan Jing, Shuyu Lin, Binyun Huang, Tingjun Luo, Xianya Li, Weiming Zhang, Shaohan Sun iD

Journal of Applied Clinical Medical Physics · 2026

DOI: 10.1002/acm2.70503 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Applied Clinical Medical Physics, 2026. Verfasst von Qiaoyan Jing und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiation Oncology, Wuming Hospital of Guangxi Medical University, Nanning, People's Republic of China. Hauptthemen laut MEDLINE: Four-Dimensional Computed Tomography, Cone-Beam Computed Tomography, Phantoms, Imaging, Radiotherapy Planning, Computer-Assisted. Die Arbeit nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Youth Science Foundation of Guangxi Medical University, Guangxi Medical and Health Appropriate Technology Development, Promotion Application Project and the Guangxi Medical and Health Self Funded Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

This study systematically quantifies the effects of five variables—respiratory cycle, tumor size, and motion amplitudes in the superior‐inferior (SI), anterior‐posterior (AP), and left‐right (LR) directions—on the registration accuracy between four‐dimensional computed tomography (4D CT) and four‐dimensional cone‐beam CT (4D CBCT) images, thereby providing a theoretical basis for optimizing registration strategies in image‐guided radiotherapy (IGRT). Materials and methods A CIRS 008A dynamic phantom fitted with 1 and 3 cm tumor inserts was utilized to simulate various respiratory motion scenarios by manipulating respiratory cycles ( T = 0, 2, 4, and 8 s) and three‐dimensional motion amplitudes (SI, AP, and LR ranging from 0 to 15 mm, with AP and LR limited to 0, 1, and 5 mm). Corresponding four‐dimensional images were acquired using a GE Discovery RT CT simulator and a Varian VitalBeam linear accelerator. Rigid registration between the 4D CT and 4D CBCT images was subsequently performed using the Varian imaging system, with registration quality quantitatively assessed via the Dice similarity coefficient (DSC). Furthermore, a Random Forest regression model was employed to determine the relative importance of each factor, and multifactor analysis of variance (ANOVA) was conducted to verify statistical significance.

Results

The Random Forest analysis indicated that, for the overall registration average intensity projection, the factors were ranked in order of importance as follows: tumor size (0.509), SI motion (0.315), respiratory cycle (0.094), LR motion (0.055), and AP motion (0.028). In the maximum intensity projection, tumor size (0.722) was found to have a particularly significant impact. The multifactor ANOVA further supported these findings, demonstrating that tumor size ( p < 0.001) and SI motion ( p < 0.001) have a highly significant influence on registration quality, whereas the respiratory cycle and AP/LR motions did not reach statistical significance ( p > 0.05). Notably, when the tumor size was small (1 cm) and accompanied by considerable SI motion (>10 mm), registration accuracy markedly deteriorated, with the greatest variability observed under these conditions.

Conclusion

This study demonstrated that the registration quality between 4D CT and 4D CBCT images was significantly influenced by both tumor size and the amplitude of motion in the SI direction.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Four-Dimensional Computed TomographyCone-Beam Computed TomographyPhantoms, ImagingRadiotherapy Planning, Computer-AssistedMachine LearningLung NeoplasmsImage Processing, Computer-AssistedRadiotherapy, Image-GuidedHumansMovementRadiotherapy, Intensity-ModulatedRespirationTumor BurdenRadiotherapy DosageAlgorithms

Stichwörter der Autor:innen

4D CBCT4D CTDice similarity coefficientimage registrationrandom forest analysis

Autor:innen und Institutionen

  • Department of Radiation Oncology, Wuming Hospital of Guangxi Medical University, Nanning, People's Republic of China.
  • Department of Radiation Oncology, Guangxi Medical University Cancer Hospital, Nanning, People's Republic of China.

ORCID-Profile: Shaohan Sun

Förderung

  • Youth Science Foundation of Guangxi Medical University
  • Guangxi Medical and Health Appropriate Technology Development
  • Promotion Application Project and the Guangxi Medical and Health Self Funded Project

Zitieren

APA

Jing, Q., Lin, S., Huang, B., Luo, T., Li, X., Zhang, W., & Sun, S. (2026). Quantifying the impact of tumor size and motion on 4DCT‐4DCBCT image registration accuracy using machine learning and statistical analysis. Journal of Applied Clinical Medical Physics. https://doi.org/10.1002/acm2.70503

Vancouver

Jing Q, Lin S, Huang B, Luo T, Li X, Zhang W, et al. Quantifying the impact of tumor size and motion on 4DCT‐4DCBCT image registration accuracy using machine learning and statistical analysis. Journal of Applied Clinical Medical Physics. 2026. doi:10.1002/acm2.70503

BibTeX
@article{Jing2026,
  author = {Jing, Qiaoyan and Lin, Shuyu and Huang, Binyun and Luo, Tingjun and Li, Xianya and Zhang, Weiming and Sun, Shaohan},
  title = {Quantifying the impact of tumor size and motion on 4DCT‐4DCBCT image registration accuracy using machine learning and statistical analysis},
  journal = {Journal of Applied Clinical Medical Physics},
  year = {2026},
  doi = {10.1002/acm2.70503}
}
Kostenlos starten →