Motion and anatomy dual aware lung ventilation imaging by integrating Jacobian map and average CT image using dual path fusion network
2025

New Method for Lung Ventilation Imaging Using CT and Deep Learning

Sample size: 66 publication Evidence: high

Author Information

Author(s): Ma Pei, Chen Zhi, Huang Yu‐Hua, Zhao Mayang, Li Wen, Li Haojiang, Cao Di, Jiang Yi‐Quan, Zhou Ta, Cai Jing, Ren Ge

Primary Institution: The Hong Kong Polytechnic University

Hypothesis

Can integrating anatomical and motion data improve the accuracy of lung ventilation imaging?

Conclusion

The study developed a dual-aware lung ventilation imaging model that significantly enhances the accuracy of ventilation estimation by incorporating motion data.

Supporting Evidence

  • The dual-aware model achieved a mean Spearman's correlation coefficient of 0.70 with reference ventilation images.
  • CTVIDual outperformed other methods in identifying low-functional lung regions.
  • The integration of motion data significantly improved the accuracy of lung ventilation estimation.

Takeaway

Researchers created a new way to see how well lungs work by combining pictures of lung anatomy with movement data, which helps doctors treat lung cancer better.

Methodology

The study used a dual-path fusion network to integrate anatomical information from CT images and motion data from Jacobian maps to synthesize ventilation images.

Potential Biases

Potential bias due to the reliance on specific datasets and the sensitivity of the model to artifacts in imaging.

Limitations

The model's accuracy may be affected by artifacts in the Jacobian maps and the limited diversity of the training dataset.

Participant Demographics

Patients included in the study were lung cancer patients with varying conditions.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1002/mp.17466

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication