Clinical Target-Volume Delineation in Prostate Brachytherapy Using Residual Neural Networks
Low dose-rate prostate brachytherapy is commonly used to treat early stage prostate cancer. This intervention involves implanting radioactive seeds inside a volume containing the prostate. Planning the intervention requires obtaining a series of ultrasound images from the prostate. This is followed by delineation of a clinical target volume, which mostly traces the prostate boundary in the ultrasound data, but can be modified based on institution-specific clinical guidelines. Here, we aim to automate the delineation of clinical target volume by using a new deep learning network based on residual neural nets and dilated convolution at deeper layers. In addition, we propose to include an exponential weight map in the optimization to improve local prediction. We train the network on 4,284 expert-labeled transrectal ultrasound images and test it on an independent set of 1,081 ultrasound images. With respect to the gold-standard delineation, we achieve a mean Dice similarity coefficient of 94%, a mean surface distance error of 1.05 mm and a mean Hausdorff distance error of 3.0 mm. The obtained results are statistically significantly better than two previous state-of-the-art techniques.
KeywordsObject segmentation Deep convolutional neural networks Residual networks Dilated convolution Clinical target volume Prostate
We would like to thank the Natural Sciences and Engineering Research Council of Canada, and the Canadian Institutes of Health Research for funding this project. The authors also would like to thank the physicians and staff at the Vancouver Cancer Centre who have contributed to this project.
- 1.Çiçek, Ö., Abdulkadir, A., Lienkamp, S.S., Brox, T., Ronneberger, O.: 3D U-Net: learning dense volumetric segmentation from sparse annotation. In: Ourselin, S., Joskowicz, L., Sabuncu, M.R., Unal, G., Wells, W. (eds.) MICCAI 2016. LNCS, vol. 9901, pp. 424–432. Springer, Cham (2016). doi: 10.1007/978-3-319-46723-8_49 CrossRefGoogle Scholar
- 2.Gibson, E., et al.: Deep residual networks for automatic segmentation of laparoscopic videos of the liver. In: SPIE Medical Imaging (2017)Google Scholar
- 3.He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: IEEE CVPR, pp. 770–778 (2016)Google Scholar
- 4.Jia, Y., et al.: Caffe: convolutional architecture for fast feature embedding. In: Proceeding International Conference Multimedia, pp. 675–678. ACM (2014)Google Scholar
- 5.Mahdavi, S.S., Chng, N., Spadinger, I., Morris, W.J., Salcudean, S.E.: Semi-automatic segmentation for prostate interventions. MIA 15(2), 226–237 (2011)Google Scholar
- 7.Nouranian, S., et al.: A multi-atlas-based segmentation framework for prostate brachytherapy. IEEE TMI 34(4), 950–961 (2015)Google Scholar
- 8.Nouranian, S., et al.: Learning-based multi-label segmentation of transrectal ultrasound images for prostate brachytherapy. IEEE TMI 35(3), 921–932 (2016)Google Scholar
- 11.Yu, F., Koltun, V.: Multi-scale context aggregation by dilated convolutions. arXiv preprint arXiv:1511.07122 (2015)