Abstract
Although the left ventricle (LV) is commonly assessed in current clinical practice, the assessment of the right ventricle (RV) also plays an important role in the diagnosis of cardiovascular disease. RV failure has numerous causes, including pulmonary hypertension, myocardial infarction, and congenital heart disease. However, assessment of the RV is more challenging than the LV due to its complex shape and thin walls. This study proposes an automated approach to delineate the RV from magnetic resonance imaging (MRI) scans using a deep convolutional neural network approach. The proposed method uses nnU-Net, a self-adapting framework based on the U-Net neural network approach for the segmentation of the RV from short and long axis MRI images at the end-systolic and end-diastolic phases of the heart. The proposed neural network models were trained using the datasets provided by Multi-Disease, Multi-View & Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge hosted by MICCAI 2021 conference. The quantitative evaluations were performed by the challenge organizers on a test set consisting of MRI scans acquired from 160 patients where the images and ground truth were blinded to the challenge participants. The proposed method yielded an overall Dice metric of \(92.47\%\) with \(92.73\%\) and \(91.71\%\) for short and long axis images, respectively. The corresponding Hausdorff distance values were 9.08 mm, 10.05 mm, and 6.16 mm, respectively.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Antonelli, M., et al.: The medical segmentation decathlon. arXiv:2106.05735, June 2021
Avendi, M.R., Kheradvar, A., Jafarkhani, H.: Automatic segmentation of the right ventricle from cardiac MRI using a learning-based approach. Magn. Reson. Med. 78(6), 2439–2448 (2017)
Bernard, O., et al.: Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved? IEEE Trans. Med. Imaging 37(11), 2514–2525 (2018)
Campello, V.M., et al.: Multi-centre, multi-vendor and multi-disease cardiac segmentation: the M&Ms challenge. IEEE Trans. Med. Imaging 40(12), 3543–3554 (2021). https://doi.org/10.1109/TMI.2021.3090082
Caudron, J., Fares, J., Lefebvre, V., Vivier, P.H., Petitjean, C., Dacher, J.N.: Cardiac MRI assessment of right ventricular function in acquired heart disease: Factors of variability. Acad. Radiol. 19(8), 991–1002 (2012)
Haddad, F., Hunt, S.A., Rosenthal, D.N., Murphy, D.J.: Right ventricular function in cardiovascular disease, Part I: anatomy, physiology, aging, and functional assessment of the right ventricle. Circulation 117(11), 1436–1448 (2008)
Isensee, F., et al.: nnU-Net: self-adapting framework for U-Net-based medical image segmentation. arXiv:1809.10486, September 2018
Mahapatra, D., Buhmann, J.M.: Automatic cardiac RV segmentation using semantic information with graph cuts. In: 2013 IEEE 10th International Symposium on Biomedical Imaging (ISBI), pp. 1106–1109. IEEE (2013)
Nambakhsh, C.M., Peters, T.M., Islam, A., Ayed, I.B.: Right ventricle segmentation with probability product kernel constraints. In: Medical Image Computing and Computer-Assisted Intervention (2013)
Petitjean, C., et al.: Right ventricle segmentation from cardiac MRI: a collation study. Med. Image Anal. 19(1), 187–202 (2015)
Punithakumar, K., Boulanger, P., Noga, M.: A GPU-accelerated deformable image registration algorithm with applications to right ventricular segmentation. IEEE Access 5, 20374–20382 (2017)
Punithakumar, K., Noga, M., Ayed, I.B., Boulanger, P.: Right ventricular segmentation in cardiac MRI with moving mesh correspondences. Comput. Med. Imaging Graph. 43, 15–25 (2015)
Punithakumar, K., Tahmasebi, N., Boulanger, P., Noga, M.: Convolutional neural network based automated RV segmentation for hypoplastic left heart syndrome MRI. In: 8th International Conference of Pattern Recognition Systems, pp. 1–6 (2017)
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Medical Image Computing and Computer-Assisted Intervention, vol. 9351, pp. 234–241 (2015). https://doi.org/10.1007/978-3-319-24574-4
Acknowledgment
The authors wish to thank the challenge organizers for providing train and test datasets as well as performing the algorithm evaluation. The authors of this paper declare that the segmentation method they implemented for participation in the M&Ms challenge has not used any pre-trained models nor additional MRI datasets other than those provided by the organizers. A. Carscadden was supported by an Undergraduate Student Research Award by the Natural Sciences and Engineering Research Council of Canada (NSERC). This research was enabled in part by computing support provided by Compute Canada (www.computecanada.ca) and WestGrid.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Punithakumar, K., Carscadden, A., Noga, M. (2022). Automated Segmentation of the Right Ventricle from Magnetic Resonance Imaging Using Deep Convolutional Neural Networks. In: Puyol Antón, E., et al. Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge. STACOM 2021. Lecture Notes in Computer Science(), vol 13131. Springer, Cham. https://doi.org/10.1007/978-3-030-93722-5_37
Download citation
DOI: https://doi.org/10.1007/978-3-030-93722-5_37
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-93721-8
Online ISBN: 978-3-030-93722-5
eBook Packages: Computer ScienceComputer Science (R0)
