EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Perfusion estimation from dynamic non-contrast computed tomography using self-supervised learning and a physics-inspired U-net transformer architecture

Yi-Kuan Liu, Jorge Cisneros, Girish Nair, Craig Stevens, Richard Castillo, Yevgeniy Vinogradskiy, Edward Castillo

International Journal of Computer Assisted Radiology and Surgery · 2025

DOI: 10.1007/s11548-025-03323-2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computer Assisted Radiology and Surgery, 2025. Verfasst von Yi-Kuan Liu und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Engineering, The University of Texas at Austin, Austin, TX, USA. Hauptthemen laut MEDLINE: Perfusion Imaging, Supervised Machine Learning, Deep Learning, Tomography, X-Ray Computed. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCI NIH HHS, NHLBI NIH HHS, NIH National Cancer Institute.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

Pulmonary perfusion imaging is a key lung health indicator with clinical utility as a diagnostic and treatment planning tool. However, current nuclear medicine modalities face challenges like low spatial resolution and long acquisition times which limit clinical utility to non-emergency settings and often placing extra financial burden on the patient. This study introduces a novel deep learning approach to predict perfusion imaging from non-contrast inhale and exhale computed tomography scans (IE-CT).

Methods

We developed a U-Net Transformer architecture modified for Siamese IE-CT inputs, integrating insights from physical models and utilizing a self-supervised learning strategy tailored for lung function prediction. We aggregated 523 IE-CT images from nine different 4DCT imaging datasets for self-supervised training, aiming to learn a low-dimensional IE-CT feature space by reconstructing image volumes from random data augmentations. Supervised training for perfusion prediction used this feature space and transfer learning on a cohort of 44 patients who had both IE-CT and single-photon emission CT (SPECT/CT) perfusion scans.

Results

Testing with random bootstrapping, we estimated the mean and standard deviation of the spatial Spearman correlation between our predictions and the ground truth (SPECT perfusion) to be 0.742 ± 0.037, with a mean median correlation of 0.792 ± 0.036. These results represent a new state-of-the-art accuracy for predicting perfusion imaging from non-contrast CT.

Conclusion

Our approach combines low-dimensional feature representations of both inhale and exhale images into a deep learning model, aligning with previous physical modeling methods for characterizing perfusion from IE-CT. This likely contributes to the high spatial correlation with ground truth. With further development, our method could provide faster and more accurate lung function imaging, potentially expanding its clinical applications beyond what is currently possible with nuclear medicine.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Perfusion ImagingSupervised Machine LearningDeep LearningTomography, X-Ray ComputedLungFour-Dimensional Computed TomographyHumansMaleFemaleMiddle Aged

Stichwörter der Autor:innen

Computed tomographyPulmonary perfusionSelf-supervised learningVision transformer

Autor:innen und Institutionen

  • Department of Biomedical Engineering, The University of Texas at Austin, Austin, TX, USA.
  • Division of Pulmonary and Critical Care, Beaumont Health, Royal Oak, MI, USA.
  • Division of Radiation Oncology, Beaumont Health, Royal Oak, MI, USA.
  • Division of Radiation Oncology, Emory University, Atlanta, GA, USA.
  • Division of Radiation Oncology, Thomas Jefferson University, Philadelphia, PA, USA.
  • Department of Biomedical Engineering, The University of Texas at Austin, Austin, TX, USA. edward.castillo@utexas.edu.

Förderung

  • NCI NIH HHS
  • NHLBI NIH HHS
  • NIH National Cancer Institute

Zitieren

APA

Liu, Y. K., Cisneros, J., Nair, G., Stevens, C., Castillo, R., Vinogradskiy, Y., & Castillo, E. (2025). Perfusion estimation from dynamic non-contrast computed tomography using self-supervised learning and a physics-inspired U-net transformer architecture. International Journal of Computer Assisted Radiology and Surgery. https://doi.org/10.1007/s11548-025-03323-2

Vancouver

Liu YK, Cisneros J, Nair G, Stevens C, Castillo R, Vinogradskiy Y, et al. Perfusion estimation from dynamic non-contrast computed tomography using self-supervised learning and a physics-inspired U-net transformer architecture. International Journal of Computer Assisted Radiology and Surgery. 2025. doi:10.1007/s11548-025-03323-2

BibTeX
@article{Liu2025,
  author = {Liu, Yi-Kuan and Cisneros, Jorge and Nair, Girish and Stevens, Craig and Castillo, Richard and Vinogradskiy, Yevgeniy and Castillo, Edward},
  title = {Perfusion estimation from dynamic non-contrast computed tomography using self-supervised learning and a physics-inspired U-net transformer architecture},
  journal = {International Journal of Computer Assisted Radiology and Surgery},
  year = {2025},
  doi = {10.1007/s11548-025-03323-2}
}
Kostenlos starten →