Artificially-generated consolidations and balanced augmentation increase performance of U-net for lung parenchyma segmentation on MR images
Open Access
- 9 May 2023
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 18 (5) , e0285378
Abstract
Purpose: To improve automated lung segmentation on 2D lung MR images using balanced augmentation and artificially-generated consolidations for training of a convolutional neural network (CNN). Materials and methods: From 233 healthy volunteers and 100 patients, 1891 coronal MR images were acquired. Of these, 1666 images without consolidations were used to build a binary semantic CNN for lung segmentation and 225 images (187 without consolidations, 38 with consolidations) were used for testing. To increase CNN performance of segmenting lung parenchyma with consolidations, balanced augmentation was performed and artificially-generated consolidations were added to all training images. The proposed CNN (CNNBal/Cons) was compared to two other CNNs: CNNUnbal/NoCons—without balanced augmentation and artificially-generated consolidations and CNNBal/NoCons—with balanced augmentation but without artificially-generated consolidations. Segmentation results were assessed using Sørensen-Dice coefficient (SDC) and Hausdorff distance coefficient. Results: Regarding the 187 MR test images without consolidations, the mean SDC of CNNUnbal/NoCons (92.1 ± 6% (mean ± standard deviation)) was significantly lower compared to CNNBal/NoCons (94.0 ± 5.3%, P = 0.0013) and CNNBal/Cons (94.3 ± 4.1%, P = 0.0001). No significant difference was found between SDC of CNNBal/Cons and CNNBal/NoCons (P = 0.54). For the 38 MR test images with consolidations, SDC of CNNUnbal/NoCons (89.0 ± 7.1%) was not significantly different compared to CNNBal/NoCons (90.2 ± 9.4%, P = 0.53). SDC of CNNBal/Cons (94.3 ± 3.7%) was significantly higher compared to CNNBal/NoCons (P = 0.0146) and CNNUnbal/NoCons (P = 0.001). Conclusions: Expanding training datasets via balanced augmentation and artificially-generated consolidations improved the accuracy of CNNBal/Cons, especially in datasets with parenchymal consolidations. This is an important step towards a robust automated postprocessing of lung MRI datasets in clinical routine.Keywords
Funding Information
- German Center for Lung Research
This publication has 27 references indexed in Scilit:
- Convolutional Neural Networks with Template-Based Data Augmentation for Functional Lung Image QuantificationAcademic Radiology, 2019
- Technical challenges of quantitative chest MRI data analysis in a large cohort pediatric studyEuropean Radiology, 2018
- A framework for Fourier‐decomposition free‐breathing pulmonary 1H MRI ventilation measurementsMagnetic Resonance in Medicine, 2018
- End-to-End Adversarial Retinal Image SynthesisIEEE Transactions on Medical Imaging, 2017
- Feasibility of quantitative regional ventilation and perfusion mapping with phase‐resolved functional lung (PREFUL) MRI in healthy volunteers and COPD, CTEPH, and CF patientsMagnetic Resonance in Medicine, 2017
- Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer LearningIEEE Transactions on Medical Imaging, 2016
- MRI of the lung (2/3). Why … when … how?Insights into Imaging, 2012
- A reproducible evaluation of ANTs similarity metric performance in brain image registrationNeuroImage, 2011
- Non‐contrast‐enhanced perfusion and ventilation assessment of the human lung by means of fourier decomposition in proton MRIMagnetic Resonance in Medicine, 2009
- Fleischner Society: Glossary of Terms for Thoracic ImagingRadiology, 2008