Abstract
Purpose
In recent years, fully convolutional networks (FCNs) have been applied to various medical image segmentation tasks. However, it is difficult to generate a large amount of high-quality annotation data to train FCNs for medical image segmentation. Thus, it is desired to achieve high segmentation performances even from incomplete training data. We aim to evaluate performance of FCNs to clean noises and interpolate labels from noisy and sparsely given label images.
Methods
To evaluate the label cleaning and propagation performance of FCNs, we used 2D and 3D FCNs to perform volumetric brain segmentation from magnetic resonance image volumes, based on network training on incomplete training datasets from noisy and sparse annotation.
Results
The experimental results using pseudo-incomplete training data showed that both 2D and 3D FCNs could provide improved segmentation results from the incomplete training data, especially by using three orthogonal annotation images for network training.
Conclusion
This paper presented a validation for label cleaning and propagation based on FCNs. FCNs might have the potential to achieve improved segmentation performances even from sparse annotation data including possible noises by manual annotation, which can be an important clue to more efficient annotation.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Withery DJ, Koles ZJ (2007) A review of medical image segmentation: methods and available software. Int J Bioelectromagn 10:125–148
Hesamian MH, Jia W, He X, Kennedy P (2019) Deep learning techniques for medical image segmentation: achievement and challenges. J Digit Imaging 32:582–596
Long J, Shelhamer E, Darrell T (2015) Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 3431–3440
Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. Med Image Comput Comput Assist Interv 9351:234–241
Çiçek Ö, Abdulkadir A, Lienkamp SS, Brox T, Ronneberger O (2016) 3D U-Net: learning dense volumetric segmentation from sparse annotation. Med Image Comput Comput Assist Interv 9901:424–432
Milletari F, Navab N, Ahmadi SA (2016) V-net: Fully convolutional neural networks for volumetric medical image segmentation. In: 4th international conference on 3D vision, pp 566–571
Feng X, Yang J, Laine AF, Angelini ED (2017) Discriminative localization in CNNs for weakly-supervised segmentation of pulmonary nodules. Med Image Comput Comput Assist Interv 10435:568–576
Baur C, Albarqouni S, Navab N (2017) Semi-supervised deep learning for fully convolutional networks. Med Image Comput Comput Assist Interv 10435:311–319
Zhao Z, Yang L, Zheng H, Guldner IH, Zhang S, Chen DZ (2018) Deep learning based instance segmentation in 3D biomedical images using weak annotation. Med Image Comput Comput Assist Interv 11073:352–360
Mirikharaji Z, Yan Y, Hamarneh G (2019) Learning to segment skin lesions from noisy annotations. Med Image Comput Comput Assist Interv 11795:207–215
Schmitz B, Hagen T, Reith W (2003) Three-dimensional true FISP for high-resolution imaging of whole brain. Eur Radiol 13:1577–1582
Zeng Q, Zhou Q, Liu Z, Li C, Ni S, Xue F (2013) Preoperative detection of the neurovascular relation ship in trigeminal neuralgia using three-dimensional fast imaging employing steady-state acquisition (FIESTA) and magnetic resonance angiography (MRA). J Clin Neurosci 20:107–111
Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv:1412.6980
Zukić D, Vicory J, McCormick M, Wisse L, Gerig G, Yushkevich P, Aylward S (2016) ND morphological contour interpolation. Insight J. http://hdl.handle.net/10380/3563
Acknowledgements
Part of this research was supported by Japan Agency for Medical Research and Development (AMED) under Grant Number JP18he1602001.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
All procedures of the study with human participants were performed in compliance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study was approved by the Ethics Committee at the University of Tokyo and Tokyo Medical and Dental University.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Sugino, T., Suzuki, Y., Kin, T. et al. Label cleaning and propagation for improved segmentation performance using fully convolutional networks. Int J CARS 16, 349–361 (2021). https://doi.org/10.1007/s11548-021-02312-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11548-021-02312-5