MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Automated Quantification of Spatially Abnormal 129-Xe MRI Ventilation and Perfusion: Implications for Lung Cancer, Asthma, and COPD Interventions

M Mcintosh*, R Eddy, J Macneil, A Matheson, G Parraga, Robarts Research Institute, Western University, London, ON, CA

Presentations

(Monday, 7/13/2020) 4:30 PM - 5:30 PM [Eastern Time (GMT-4)]

Room: Track 1

Purpose: methods for quantifying ventilation and ventilation abnormalities in patients with lung cancer, asthma or chronic obstructive pulmonary disease typically measure inhaled hyperpolarized ¹²?Xe MRI signal voids. Until now, such signal voids have been measured using signal intensity histogram cluster analysis and the volume of such signal voids are normalized to the volume of the thoracic cavity as ventilation defect percent (VDP). Unfortunately, this approach disregards signal intensity inhomogeneities and their spatial distribution and therefore, much of the spatial and signal intensity information is lost. To regionally guide pulmonary interventions including radiation therapy, this quantitative information is important. Therefore, our objective was to develop a novel software pipeline to quantify the distribution of both ¹²?Xe MRI ventilation and perfusion abnormalities and defects.


Methods: ¹²?Xe MRI ventilation and perfusion maps and volume-matched ¹H MRI were acquired as previously described and used as pipeline inputs for automated co-registration and segmentation. Ventilation and perfusion MRI signal intensity histograms were generated and the signal intensity values for specific percentiles (PN) including N=5, N=10 and N=95 were calculated. We defined the area under the curve for the PN-percentile non-linear relationship as signal intensity inhomogeneity (S?). Relationships for S? and pulmonary function were evaluated using linear regression. Preliminary pipeline testing was performed using ¹²?Xe MRI acquired in 29 participants with inflammatory airways disease and abnormal pulmonary function.


Results: automated pipeline required <1s for segmentation and registration and <10s for histogram analysis and output file production. Mean S? was 24±8 (range=15-50). P5, P10, and S? but not P95 showed significant and moderate to strong relationships with VDP (r=-0.67, -0.71, -0.62; all, p<0.001) and FEV1 (r=0.60, 0.63, 0.61; all, p=0.001).


Conclusion: quantified spatially abnormal ventilation and perfusion using ¹²?Xe MRI signal intensity distributions to spatially guide interventions in patients with inflammatory airways disease.

Keywords

Lung, Image Analysis, Quantitative Imaging

Taxonomy

IM/TH- Image Analysis (Single Modality or Multi-Modality): Quantitative imaging

Contact Email