Implementable Deep Learning for Multi‐sequence Proton <scp>MRI</scp> Lung Segmentation: A Multi‐center, Multi‐vendor, and Multi‐disease Study
Joshua Astley, Alberto Biancardi, Paul Hughes, Helen Marshall, Guilhem Collier, Ho‐Fung Chan, Laura Saunders, Laurie Smith, Martin Brook, A. A. Roger Thompson, Sarah Rowland‐Jones, Sarah Skeoch, Stephen Bianchi, M. Hatton, Najib M. Rahman, Ling‐Pei Ho, Christopher E. Brightling, Louise V. Wain, Amisha Singapuri, Rachael A Evans, Alastair J. Moss, Gerry P McCann, Stefan Neubauer, Betty Raman, C‐MORE/PHOSP‐COVID Collaborative Group, Jim M. Wild, Bilal Tahir
Abstract
BACKGROUND: H)-MRI lung segmentation. However, previous deep learning studies have utilized single-center data and limited acquisition parameters. PURPOSE: H-MRI, robust to pathology, acquisition protocol, vendor, and center. STUDY TYPE: Retrospective. POPULATION: H-MRI scans from 258 participants with various pulmonary pathologies (median age (range): 57 (6-85); 42% females) and 31 healthy participants (median age (range): 34 (23-76); 34% females) that were split into training (593 scans (74%); 157 participants (55%)), testing (50 scans (6%); 50 participants (17%)) and external validation (164 scans (20%); 82 participants (28%)) sets. FIELD STRENGTH/SEQUENCE: H-MRI. ASSESSMENT: 2D and 3D CNNs, trained on single-center, multi-sequence data, and the conventional spatial fuzzy c-means (SFCM) method were compared to manually delineated expert segmentations. Each method was validated on external data originating from several centers. Dice similarity coefficient (DSC), average boundary Hausdorff distance (Average HD), and relative error (XOR) metrics to assess segmentation performance. STATISTICAL TESTS: Kruskal-Wallis tests assessed significances of differences between acquisitions in the testing set. Friedman tests with post hoc multiple comparisons assessed differences between the 2D CNN, 3D CNN, and SFCM. Bland-Altman analyses assessed agreement with manually derived lung volumes. A P value of <0.05 was considered statistically significant. RESULTS: The 3D CNN significantly outperformed its 2D analog and SFCM, yielding a median (range) DSC of 0.961 (0.880-0.987), Average HD of 1.63 mm (0.65-5.45) and XOR of 0.079 (0.025-0.240) on the testing set and a DSC of 0.973 (0.866-0.987), Average HD of 1.11 mm (0.47-8.13) and XOR of 0.054 (0.026-0.255) on external validation data. DATA CONCLUSION: H-MRI lung segmentations on a heterogenous dataset, demonstrating robustness to disease pathology, sequence, vendor, and center. EVIDENCE LEVEL: 4. TECHNICAL EFFICACY: Stage 1.