We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
A Dual‐Channel Deep Learning Approach for Lung Cavity Estimation From Hyperpolarized Gas and Proton MRI.
- Authors
Astley, Joshua R.; Biancardi, Alberto M.; Marshall, Helen; Hughes, Paul J. C.; Collier, Guilhem J.; Smith, Laurie J.; Eaden, James A.; Hughes, Rod; Wild, Jim M.; Tahir, Bilal A.
- Abstract
Background: Hyperpolarized gas MRI can quantify regional lung ventilation via biomarkers, including the ventilation defect percentage (VDP). VDP is computed from segmentations derived from spatially co‐registered functional hyperpolarized gas and structural proton (1H)‐MRI. Although acquired at similar lung inflation levels, they are frequently misaligned, requiring a lung cavity estimation (LCE). Recently, single‐channel, mono‐modal deep learning (DL)‐based methods have shown promise for pulmonary image segmentation problems. Multichannel, multimodal approaches may outperform single‐channel alternatives. Purpose: We hypothesized that a DL‐based dual‐channel approach, leveraging both 1H‐MRI and Xenon‐129‐MRI (129Xe‐MRI), can generate LCEs more accurately than single‐channel alternatives. Study Type: Retrospective. Population: A total of 480 corresponding 1H‐MRI and 129Xe‐MRI scans from 26 healthy participants (median age [range]: 11 [8–71]; 50% females) and 289 patients with pulmonary pathologies (median age [range]: 47 [6–83]; 51% females) were split into training (422 scans [88%]; 257 participants [82%]) and testing (58 scans [12%]; 58 participants [18%]) sets. Field Strength/Sequence: 1.5‐T, three‐dimensional (3D) spoiled gradient‐recalled 1H‐MRI and 3D steady‐state free‐precession 129Xe‐MRI. Assessment: We developed a multimodal DL approach, integrating 129Xe‐MRI and 1H‐MRI, in a dual‐channel convolutional neural network. We compared this approach to single‐channel alternatives using manually edited LCEs as a benchmark. We further assessed a fully automatic DL‐based framework to calculate VDPs and compared it to manually generated VDPs. Statistical Tests: Friedman tests with post hoc Bonferroni correction for multiple comparisons compared single‐channel and dual‐channel DL approaches using Dice similarity coefficient (DSC), average boundary Hausdorff distance (average HD), and relative error (XOR) metrics. Bland–Altman analysis and paired t‐tests compared manual and DL‐generated VDPs. A P value < 0.05 was considered statistically significant. Results: The dual‐channel approach significantly outperformed single‐channel approaches, achieving a median (range) DSC, average HD, and XOR of 0.967 (0.867–0.978), 1.68 mm (37.0–0.778), and 0.066 (0.246–0.045), respectively. DL‐generated VDPs were statistically indistinguishable from manually generated VDPs (P = 0.710). Data Conclusion: Our dual‐channel approach generated LCEs, which could be integrated with ventilated lung segmentations to produce biomarkers such as the VDP without manual intervention. Evidence Level: 4. Technical Efficacy: Stage 1.
- Subjects
CONVOLUTIONAL neural networks; DEEP learning; MAGNETIC resonance imaging; PROTONS; SIGNAL convolution; IMAGE segmentation; BLAND-Altman plot
- Publication
Journal of Magnetic Resonance Imaging, 2023, Vol 57, Issue 6, p1878
- ISSN
1053-1807
- Publication type
Article
- DOI
10.1002/jmri.28519