Quantification of ¹²⁹Xe MRI Ventilation-defect-percent Using Binary-threshold, Gaussian Linear-Binning and K-means...

Area of Science:

• Pulmonary Medicine
• Radiology
• Medical Imaging

Background:

• Hyperpolarized 129Xe MRI quantifies ventilation heterogeneity using ventilation defect percent (VDP).
• Current VDP quantification methods include binary threshold (BT), Gaussian-linear-binning (GLB), and k-means clustering.
• These methods have not been directly compared in asthma and COPD patients, where ventilation defects are common.

Purpose of the Study:

• To quantify and compare VDP using four distinct methods in patients with asthma and COPD.
• To evaluate the agreement and correlation between different VDP quantification techniques.

Main Methods:

• Retrospective analysis of data from 175 participants (42 healthy, 43 COPD, 90 asthma).
• VDP was quantified using CNN co-registration and segmentation, GLB, GLBslice, BT, and k-means methods.
• Linear regression and Bland-Altman plots were employed to assess inter-method correlations and agreement.

Main Results:

• Significant differences in VDP were observed between GLB/BT and GLBslice/k-means methods across all groups.
• VDP calculated using GLB, GLBslice, and BT showed significant correlations with k-means VDP (R² values ranging from 0.64 to 0.84).
• Bland-Altman analysis revealed wide 95% confidence intervals for agreement between k-means and the other methods, particularly in COPD patients.

Conclusions:

• The choice of VDP quantification method significantly impacts results in hyperpolarized 129Xe MRI studies of asthma and COPD.
• Discrepancies observed between methods necessitate careful consideration for standardization and interpretation in multi-center research.
• Further investigation into method harmonization is crucial for reliable clinical application and comparative studies.