Publications
Take a look

by Joshua R. Astley BEng, Alberto M. Biancardi PhD, Paul J. C. Hughes PhD, Helen Marshall PhD, Guilhem J. Collier PhD, Ho-Fung Chan PhD, Laura C. Saunders PhD, Laurie J. Smith PhD, Martin L. Brook MSc, Roger Thompson PhD, Sarah Rowland-Jones MD, Sarah Skeoch PhD, Stephen M. Bianchi PhD, Matthew Q. Hatton MD, Najib M. Rahman DPhil, Ling-Pei Ho PhD, Chris E. Brightling PhD, Louise V. Wain PhD, Amisha Singapuri BSc, Rachael A. Evans PhD, Alastair J. Moss PhD, Gerry P. McCann MD, Stefan Neubauer MD, Betty Raman DPhil, C-MORE/PHOSP-COVID Collaborative Group, Jim M. Wild PhD, Bilal A. Tahir PhD on behalf of the TRISTAN Consortium

 Journal of Magnetic Resonance Imaging 58, Nr. 4 (Oktober 2023): 1030–44. doi: 10.1002/jmri.28643

Abstract

Background
Recently, deep learning via convolutional neural networks (CNNs) has largely superseded conventional methods for proton (1H)-MRI lung segmentation. However, previous deep learning studies have utilized single-center data and limited acquisition parameters.

Purpose
Develop a generalizable CNN for lung segmentation in 1H-MRI, robust to pathology, acquisition protocol, vendor, and center.

Study type
Retrospective.

Population
A total of 809 1H-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
1.5-T and 3-T/3D spoiled-gradient recalled and ultrashort echo-time 1H-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
The 3D CNN generated accurate 1H-MRI lung segmentations on a heterogenous dataset, demonstrating robustness to disease pathology, sequence, vendor, and center.