EUVIMEDTutustu EUVIMEDiin
Open access · CC BY 4.0

Motion and anatomy dual aware lung ventilation imaging by integrating Jacobian map and average CT image using dual path fusion network

Pei Ma, Zhi Chen, Yu-Hua Huang, Mayang Zhao, Wen Li, Haojiang Li, Di Cao, Yi-Quan Jiang, Ta Zhou, Jing Cai, Ge Ren

Medical Physics · 2025

DOI: 10.1002/mp.17466 ↗

At a glance

Journal article in Medical Physics, 2025. Written by Pei Ma and 10 co-authors. Institution: Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Kowloon, Hong Kong SAR, Hong Kong. Major MEDLINE topics: Four-Dimensional Computed Tomography, Lung, Image Processing, Computer-Assisted. Cited 5 times so far (Crossref); 44 references. The full text is freely available (open access, CC BY 4.0). Funded by: Health and Medical Research Fund, Health Bureau, the Pneumoconiosis Compensation Fund Board in HKSAR, and Shenzhen Science and Technology Program, the Shenzhen Basic Research Program.

TiivistelmäSubjectsBibliografiset tiedotCiteRelated studies

Abstract (original language)

Background

Deep learning‐based computed tomography (CT) ventilation imaging (CTVI) is a promising technique for guiding functional lung avoidance radiotherapy (FLART). However, conventional approaches, which rely on anatomical CT data, may overlook important ventilation features due to the lack of motion data integration.

Purpose

This study aims to develop a novel dual‐aware CTVI method that integrates both anatomical information from CT images and motional information from Jacobian maps to generate more accurate ventilation images for FLART.

Methods

A dataset of 66 patients with four‐dimensional CT (4DCT) images and reference ventilation images (RefVI) was utilized to develop the dual‐path fusion network (DPFN) for synthesizing ventilation images (CTVI Dual ). The DPFN model was specifically designed to integrate motion data from 4DCT‐generated Jacobian maps with anatomical data from average 4DCT images. The DPFN utilized two specialized feature extraction pathways, along with encoders and decoders, designed to handle both 3D average CT images and Jacobian map data. This dual‐processing approach enabled the comprehensive extraction of lung ventilation‐related features. The performance of DPFN was assessed by comparing CTVI Dual to RefVI using various metrics, including Spearman's correlation coefficients ( R ), Dice similarity coefficients of high‐functional region (DSC h ), and low‐functional region (DSC l ). Additionally, CTVI Dual was benchmarked against other CTVI methods, including a dual‐phase CT‐based deep learning method (CTVI DLCT ), a radiomics‐based method (CTVI FM ), a super voxel‐based method (CTVI SVD ), a Unet‐based method (CTVI Unet ), and two deformable registration‐based methods (CTVI Jac and CTVI HU ).

Results

In the test group, the mean R between CTVI Dual and RefVI was 0.70, significantly outperforming CTVI DLCT (0.68), CTVI FM (0.58), CTVI SVD (0.62), and CTVI Unet (0.66), with p < 0.05. Furthermore, the DSC h and DSC l values of CTVI Dual were 0.64 and 0.80, respectively, outperforming CTVI SVD (0.63; 0.73) and CTVI Unet (0.62; 0.77). The performance of CTVI Dual was also significantly better than that of CTVI Jac and CTVI HU .

Conclusions

A novel dual‐aware CTVI model that integrates anatomical and motion information was developed to synthesize lung ventilation images. It was shown that the accuracy of lung ventilation estimation could be significantly enhanced by incorporating motional information, particularly in patients with tumor‐induced blockages. This approach has the potential to improve the accuracy of CTVI, enabling more effective FLART.

Subjects

MeSH terms (MEDLINE)

Four-Dimensional Computed TomographyLungImage Processing, Computer-AssistedHumansMovementDeep LearningPulmonary VentilationRespirationLung Neoplasms

Author keywords

CT ventilation imagingJacobian mapdeep learningfunctional lung avoidance radiotherapy

Authors and institutions

  • Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Kowloon, Hong Kong SAR, Hong Kong.
  • Department of Radiology, State Key Laboratory of Oncology in South China, Collaborative Innovation Centre for Cancer Medicine, Guangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and Therapy, Sun Yat-sen Univ…
  • Department of Minimally Invasive Interventional Therapy, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou, People…
  • The Hong Kong Polytechnic University Shenzhen Research Institute, Shenzhen, Guangdong Province, People's Republic of China.

Funding

  • Health and Medical Research Fund
  • Health Bureau, the Pneumoconiosis Compensation Fund Board in HKSAR, and Shenzhen Science and Technology Program
  • the Shenzhen Basic Research Program

Cite

APA

Ma, P., Chen, Z., Huang, Y. H., Zhao, M., Li, W., Li, H., Cao, D., Jiang, Y. Q., Zhou, T., Cai, J., & Ren, G. (2025). Motion and anatomy dual aware lung ventilation imaging by integrating Jacobian map and average CT image using dual path fusion network. Medical Physics. https://doi.org/10.1002/mp.17466

Vancouver

Ma P, Chen Z, Huang YH, Zhao M, Li W, Li H, et al. Motion and anatomy dual aware lung ventilation imaging by integrating Jacobian map and average CT image using dual path fusion network. Medical Physics. 2025. doi:10.1002/mp.17466

BibTeX
@article{Ma2025,
  author = {Ma, Pei and Chen, Zhi and Huang, Yu-Hua and Zhao, Mayang and Li, Wen and Li, Haojiang and Cao, Di and Jiang, Yi-Quan and Zhou, Ta and Cai, Jing and Ren, Ge},
  title = {Motion and anatomy dual aware lung ventilation imaging by integrating Jacobian map and average CT image using dual path fusion network},
  journal = {Medical Physics},
  year = {2025},
  doi = {10.1002/mp.17466}
}
Start for free →