Journal
MAGNETIC RESONANCE IMAGING
Volume 92, Issue -, Pages 140-149Publisher
ELSEVIER SCIENCE INC
DOI: 10.1016/j.mri.2022.06.016
Keywords
MRI; Deep learning; COPD; Ventilation defects; Hyperpolarized gas
Funding
- National Heart, Lung, and Blood Institute [R01-HL093081, R01-HL077612, R01-HL121270]
- National Center for Advancing Translational Sciences (NCATS) [75N92020D00001, HHSN268201500003I, N01-HC-95169]
- National Institute of Diabetes and Digestive and Kidney Diseases [UL1-TR-000040, UL1-TR-001079, UL1-TR-001420]
- Medical Research Council [R01-HL093081]
- [P30 DK26687]
- [MR/M008894/1]
Ask authors/readers for more resources
The purpose of this study was to develop a deep learning framework for segmenting ventilation defects on pulmonary hyperpolarized MRI. Traditional U-Net and cascaded U-Net algorithms were used for segmentation, and conventional data augmentation was employed to improve the segmentation results.
Purpose: To develop an end-to-end deep learning (DL) framework to segment ventilation defects on pulmonary hyperpolarized MRI. Materials and methods: The Multi-Ethnic Study of Atherosclerosis Chronic Obstructive Pulmonary Disease (COPD) study is a nested longitudinal case-control study in older smokers. Between February 2016 and July 2017, 56 participants (age, mean +/- SD, 74 +/- 8 years; 34 men) underwent same breath-hold proton (1H) and helium (3He) MRI, which were annotated for non-ventilated, hypo-ventilated, and normal-ventilated lungs. In this retrospective DL study, 820 1H and 3He slices from 42/56 (75%) participants were randomly selected for training, with the remaining 14/56 (25%) for test. Full lung masks were segmented using a traditional U-Net on 1H MRI and were imported into a cascaded U-Net, which were used to segment ventilation defects on 3He MRI. Models were trained with conventional data augmentation (DA) and generative adversarial networks (GAN)-DA. Results: Conventional-DA improved 1H and 3He MRI segmentation over the non-DA model (P = 0.007 to 0.03) but GAN-DA did not yield further improvement. The cascaded U-Net improved non-ventilated lung segmentation (P < 0.005). Dice similarity coefficients (DSC) between manually and DL-segmented full lung, non-ventilated, hypo-ventilated, and normal-ventilated regions were 0.965 +/- 0.010, 0.840 +/- 0.057, 0.715 +/- 0.175, and 0.883 +/- 0.060, respectively. We observed no statistically significant difference in DCSs between participants with and without COPD (P = 0.41, 0.06, and 0.18 for non-ventilated, hypo-ventilated, and normal-ventilated regions, respectively).
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available