IM-Net: Semantic Segmentation Algorithm for Medical Images Based on Mutual Information Maximization
Authors: 
Yi Sun, Peisen YuanAuthors Info & Claims
Pages 397 - 405
Abstract
The medical image is often noisy, which makes it difficult to extract the image features from the medical image segmentation model. Because the noise is often generated randomly, it is difficult to use supervised information for denoising. In this paper, we focus on this challenging problem and propose an IM-Net algorithm for medical image segmentation based on mutual information maximization. The IM-Net can remove the noise and therefore improve the quality of the extracted feature by maximizing the mutual information between the extracted feature and the input image. IM-Net uses the Binary Cross Entropy with Logits Estimation to approach the true value of mutual information and uses a bilinear interpolation function as a discriminator to maximize the mutual information estimator. Extensive experiments are conducted and the IM-Net is compared with different methods to demonstrate the effectiveness of our model. Experimental results show that the training efficiency and segmentation precision are greatly improved.
References
[1]
Shen D and Ip HHS Discriminative wavelet shape descriptors for recognition of 2-D patterns Pattern Recogn. 1999 32 2 151-165
[2]
Tamura H, Mori S, and Yamawaki T Textural features corresponding to visual perception IEEE Trans. Syst. Man Cybern. 1978 8 6 460-473
[3]
Hong Z-Q Algebraic feature extraction of image for recognition Pattern Recogn. 1991 24 3 211-219
[4]
Goupillaud P, Grossmann A, and Morlet J Cycle-octave and related transforms in seismic signal analysis Geoexploration 1984 23 1 85-102
[5]
Hjelm, R.D., et al.: Learning deep representations by mutual information estimation and maximization. arXiv preprint arXiv:1808.06670 (2018)
[6]
Yang, S., Ramanan, D.: Multi-scale recognition with DAG-CNNs. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1215–1223 (2015)
[7]
Herranz, L., Jiang, S., Li, X.: Scene recognition with CNNs: objects, scales and dataset bias. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 571–579 (2016)
[8]
Gao, H., Ji, S.: Graph u-nets. In: Proceedings of the 36th International Conference on Machine Learning, ICML 2019, vol. 97, pp. 2083–2092. PMLR (2019)
[9]
Yoo, D., Park, S., Lee, J.-Y., Kweon, I.S.: Multi-scale pyramid pooling for deep convolutional representation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 71–80 (2015)
[10]
Chen C, Liu X, Ding M, Zheng J, Li J, et al. Shen D et al. 3D dilated multi-fiber network for real-time brain tumor segmentation in MRI Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 2019 Cham Springer 184-192
[11]
Huang C, Han H, Yao Q, Zhu S, and Zhou SK Shen D, Liu T, Peters TM, Staib LH, Essert C, Zhou S, Yap P-T, and Khan A 3D U-net: a 3D universal U-net for multi-domain medical image segmentation Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 2019 Cham Springer 291-299
[12]
Brügger R, Baumgartner CF, and Konukoglu E Shen D, Liu T, Peters TM, Staib LH, Essert C, Zhou S, Yap P-T, and Khan A A partially reversible U-net for memory-efficient volumetric image segmentation Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 2019 Cham Springer 429-437
[13]
Orlando, J.I., et al.: U2-net: a Bayesian U-net model with epistemic uncertainty feedback for photoreceptor layer segmentation in pathological OCT scans. In: 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019, pp. 1441–1445. IEEE (2019)
[14]
Ronneberger O, Fischer P, and Brox T Navab N, Hornegger J, Wells WM, and Frangi AF U-net: convolutional networks for biomedical image segmentation Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015 2015 Cham Springer 234-241
[15]
Zhou Z, Rahman Siddiquee MM, Tajbakhsh N, Liang J, et al. Stoyanov D et al. UNet++: a nested U-net architecture for medical image segmentation Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support 2018 Cham Springer 3-11
[16]
Milletari, F., Navab, N., Ahmadi, S.: V-net: fully convolutional neural networks for volumetric medical image segmentation. In: 2016 Fourth International Conference on 3D Vision (3DV), pp. 565–571, October 2016
[17]
Oktay, O., et al.: Attention U-net: learning where to look for the pancreas. arXiv preprint arXiv:1804.03999, 2018
[18]
Jin, Q., Meng, Z., Sun, C., Wei, L., Su, R.: RA-UNET: a hybrid deep attention-aware network to extract liver and tumor in CT scans. arXiv preprint arXiv:1811.01328 (2018)
[19]
Belghazi, M.I., et al.: Mutual information neural estimation. International Conference on Machine Learning, pp. 531–540 (2018)
Recommendations
A novel segmentation model for medical images with intensity inhomogeneity based on adaptive perturbation
In medical field, it remains challenging to accurately segment medical images due to low contrast, complex noises and intensity inhomogeneity. To overcome these obstacles, this paper provides a novel edge-based active contour model (ACM) for medical ...
A robust approach for subject segmentation of medical Images
A novel, robust technique for medical image segmentation is presented.The proposed method improves on existing methods and preserves their advantages.Real medical image results illustrate the proposed method's excellent performance.Comparison with ...
Incorporating priors for medical image segmentation using a genetic algorithm
Medical image segmentation is typically performed manually by a physician to delineate gross tumor volumes for treatment planning and diagnosis. Manual segmentation is performed by medical experts using prior knowledge of organ shapes and locations but ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
Aug 2020
524 pages
ISBN:978-3-030-55129-2
DOI:10.1007/978-3-030-55130-8
© Springer Nature Switzerland AG 2020.
Publisher
Springer-Verlag
Berlin, Heidelberg
Publication History
Published: 28 August 2020
Author Tags
Qualifiers
- Article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 16 Dec 2024