Author: Fei Han, James M. Lamb, Michael Vincent Lauria, Daniel A. Low, Tessa Elizabeth Maurer, Danilo Maziero, Claudia R. Miller, Drew Moghanaki, Louise Naumann, Nicolas Viot 👨🔬
Affiliation: Department of Radiation Oncology, University of California, Los Angeles, Siemens Healthineers, UCLA, University of California Los Angeles 🌍
Purpose: Patients with lung disease outside radiotherapy are barred from high dose protocols used for motion modeling, but MRI could offer no-dose alternatives. Image-based ventilation is a promising technique for assessing lung function. This study evaluates end-exhalation to end-inhalation deformable image registration (DIR) derived from free-breathing MRI lung scans subsequently used in ventilation calculations to evaluate the feasibility of an MR-based approach.
Methods: For a cohort of five volunteers (4 healthy; 1 patient with lung cancer), free-breathing lung scans were acquired on a 1.5T MRI Siemens Magnetom Sola. An Ultra-Short Echo Time stack-of-spirals sequence, with Reconstructed Golden-Angle Radial Sparse Parallel, was used to account for respiratory motion focusing on end-exhalation and end-inhalation. The DEnse Displacement Sampling (deeds) DIR algorithm was used to determine the end-inhale to end-exhale deformation vector field (DVF). Following AAPM’s TG-132, the registrations were evaluated with target registration error (TRE) for five bifurcation landmarks across the upper and lower lobes of each lung. Additionally, the determinant of the Jacobian matrix of the DVF provided ventilation mapping.
Results: The mean TRE was 2.1±1.6mm, which is slightly larger than the voxel size (1.5mm) but smaller than the bifurcation features. The Jacobian determinant was AAPM TG-132 compliant as all volunteers had positive Jacobian determinants. The mean lung ventilation across volunteers was 1.17±0.16, which is consistent with healthy subjects.
Conclusion: This work indicates that DIR of free-breathing MRI lung scans are feasible according to metrics outlined in AAPM’s TG-132. With accurate registration, we can further develop a ventilation technique for lung disease patients outside radiotherapy. Future work will expand the number of TRE landmarks and quantify additional DIR metrics such as mean distance to agreement and dice similarity coefficient.