Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Apr:2018:806-810.
doi: 10.1109/ICASSP.2018.8461716. Epub 2018 Sep 13.

A NOVEL IMAGE-SPECIFIC TRANSFER APPROACH FOR PROSTATE SEGMENTATION IN MR IMAGES

Pinzhuo Tian  1 , Lei Qi  1 , Yinghuan Shi  1 , Luping Zhou  2 , Yang Gao  1 , Dinggang Shen  3
Affiliations

A NOVEL IMAGE-SPECIFIC TRANSFER APPROACH FOR PROSTATE SEGMENTATION IN MR IMAGES

Pinzhuo Tian et al. Proc IEEE Int Conf Acoust Speech Signal Process. 2018 Apr.

Abstract

Prostate segmentation in Magnetic Resonance (MR) Images is a significant yet challenging task for prostate cancer treatment. Most of the existing works attempted to design a global classifier for all MR images, which neglect the discrepancy of images across different patients. To this end, we propose a novel transfer approach for prostate segmentation in MR images. Firstly, an image-specific classifier is built for each training image. Secondly, a pair of dictionaries and a mapping matrix are jointly obtained by a novel Semi-Coupled Dictionary Transfer Learning (SCDTL). Finally, the classifiers on the source domain could be selectively transferred to the target domain (i.e. testing images) by the dictionaries and the mapping matrix. The evaluation demonstrates that our approach has a competitive performance compared with the state-of-the-art transfer learning methods. Moreover, the proposed transfer approach outperforms the conventional deep neural network based method.

Keywords: Image-Specific Transfer Approach; Prostate Segmentation; Semi-Coupled Dictionary Transfer Learning.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
The framework of the proposed method for prostate segmentation in MR images.
Fig. 2
Fig. 2
The visualized results of the proposed method and baselines. Note that the red line represents groundtruth and the green line denotes segmentation result.
See this image and copyright information in PMC

References

    1. Yu Lequan, Yang Xin, Chen Hao, Qin Jing, Heng Pheng-Ann. Volumetric convnets with mixed residual connections for automated prostate segmentation from 3d mr images. AAAI. 2017:66–72.
    1. Guo Yanrong, Gao Yaozong, Shen Dinggang. Deep Learning for Medical Image Analysis. Elsevier; 2017. Deformable mr prostate segmentation via deep feature learning and sparse patch matching; pp. 197–222. - PMC - PubMed
    1. Clark Tyler, Wong Alexander, Haider Masoom A, Khalvati Farzad. International Conference Image Analysis and Recognition. Springer; 2017. Fully deep convolutional neural networks for segmentation of the prostate gland in diffusion-weighted mr images; pp. 97–104.
    1. Tian Zhiqiang, Liu LiZhi, Fei Baowei. Medical Imaging 2015: Image Processing. Vol. 9413. International Society for Optics and Photonics; 2015. A supervoxel-based segmentation method for prostate mr images; p. 941318. - PMC - PubMed
    1. Klein Stefan, Van Der Heide Uulke A, Lips Irene M, Vulpen Marco Van, Staring Marius, Pluim Josien PW. Automatic segmentation of the prostate in 3d mr images by atlas matching using localized mutual information. Medical physics. 2008;35(4):1407–1417. - PubMed

LinkOut - more resources