Deep convolutional neural networks with multiplane consensus labeling for lung function quantification using UTE proton MRI

Wei Zha, Sean B. Fain, Mark L. Schiebler, Michael D. Evans, Scott K. Nagle, Fang Liu

Research output: Contribution to journalArticlepeer-review

22 Scopus citations


BACKGROUND: Ultrashort echo time (UTE) proton MRI has gained popularity for assessing lung structure and function in pulmonary imaging; however, the development of rapid biomarker extraction and regional quantification has lagged behind due to labor-intensive lung segmentation.

PURPOSE: To evaluate a deep learning (DL) approach for automated lung segmentation to extract image-based biomarkers from functional lung imaging using 3D radial UTE oxygen-enhanced (OE) MRI.

STUDY TYPE: Retrospective study aimed to evaluate a technical development.

POPULATION: Forty-five human subjects, including 16 healthy volunteers, 5 asthma, and 24 patients with cystic fibrosis.

FIELD STRENGTH/SEQUENCE: 1.5T MRI, 3D radial UTE (TE = 0.08 msec) sequence.

ASSESSMENT: Two 3D radial UTE volumes were acquired sequentially under normoxic (21% O 2 ) and hyperoxic (100% O 2 ) conditions. Automated segmentation of the lungs using 2D convolutional encoder-decoder based DL method, and the subsequent functional quantification via adaptive K-means were compared with the results obtained from the reference method, supervised region growing.

STATISTICAL TESTS: Relative to the reference method, the performance of DL on volumetric quantification was assessed using Dice coefficient with 95% confidence interval (CI) for accuracy, two-sided Wilcoxon signed-rank test for computation time, and Bland-Altman analysis on the functional measure derived from the OE images.

RESULTS: The DL method produced strong agreement with supervised region growing for the right (Dice: 0.97; 95% CI = [0.96, 0.97]; P < 0.001) and left lungs (Dice: 0.96; 95% CI = [0.96, 0.97]; P < 0.001). The DL method averaged 46 seconds to generate the automatic segmentations in contrast to 1.93 hours using the reference method (P < 0.001). Bland-Altman analysis showed nonsignificant intermethod differences of volumetric (P ≥ 0.12) and functional measurements (P ≥ 0.34) in the left and right lungs.

DATA CONCLUSION: DL provides rapid, automated, and robust lung segmentation for quantification of regional lung function using UTE proton MRI.

LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2019;50:1169-1181.

Original languageEnglish (US)
Pages (from-to)1169-1181
Number of pages13
JournalJournal of Magnetic Resonance Imaging
Issue number4
StatePublished - Oct 1 2019

Bibliographical note

Funding Information:
Contract grant sponsor: NIH/NCATS; Contract grant number: UL1TR000427; Contract grant sponsor: NIH/NHLBI; Contract grant number: U10 HL109168; Contract grant sponsor: ICTR; Contract grant numbers: NIH R01AR068373, NIH/NHLBI, R01 HL126771; Contract grant sponsor: Research and Development Fund from the Departments of Radiology and Medical Physics, University of Wisconsin-Madison. The authors thank Robert V. Cadman, PhD, who administered the gas and facemask supplies and advised on the experimental setup; the nurses and recruiters who supported this work, especially Jan Yakey, RN, and Gina Crisafi, BS; and the research technologists at University of Wisconsin-Madison, including Kelli HellenBrand, RT, Jenelle Fuller, RT, and Sara John, RT, who supported the MRI scanning.

Publisher Copyright:
© 2019 International Society for Magnetic Resonance in Medicine


  • asthma
  • cystic fibrosis
  • deep learning
  • hyperoxia
  • lung
  • magnetic resonance imaging

PubMed: MeSH publication types

  • Journal Article


Dive into the research topics of 'Deep convolutional neural networks with multiplane consensus labeling for lung function quantification using UTE proton MRI'. Together they form a unique fingerprint.

Cite this