Biomedical image segmentation algorithm based on dense atrous convolution

Hong'an Li; Man Liu; Jiangwen Fan; Qingfang Liu; Hong'an Li; Man Liu; Jiangwen Fan; Qingfang Liu

doi:10.3934/mbe.2024192

Mathematical Biosciences and Engineering

2024, Volume 21, Issue 3: 4351-4369. doi: 10.3934/mbe.2024192

Previous Article Next Article

Research article Special Issues

Biomedical image segmentation algorithm based on dense atrous convolution

1.
College of Computer Science and Technology, Xi'an University of Science and Technology, Xi'an, 710054, China
2.
Information Center, Shijiazhuang Posts and Telecommunications Technical College, Shijiazhuang 050021, China
3.
State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, Beijing, 100191, China

Received: 19 November 2023 Revised: 15 February 2024 Accepted: 20 February 2024 Published: 26 February 2024

Biomedical images have complex tissue structures, and there are great differences between images of the same part of different individuals. Although deep learning methods have made some progress in automatic segmentation of biomedical images, the segmentation accuracy is relatively low for biomedical images with significant changes in segmentation targets, and there are also problems of missegmentation and missed segmentation. To address these challenges, we proposed a biomedical image segmentation method based on dense atrous convolution. First, we added a dense atrous convolution module (DAC) between the encoding and decoding paths of the U-Net network. This module was based on the inception structure and atrous convolution design, which can effectively capture multi-scale features of images. Second, we introduced a dense residual pooling module to detect multi-scale features in images by connecting residual pooling blocks of different sizes. Finally, in the decoding part of the network, we adopted an attention mechanism to suppress background interference by enhancing the weight of the target area. These modules work together to improve the accuracy and robustness of biomedical image segmentation. The experimental results showed that compared to mainstream segmentation networks, our segmentation model exhibited stronger segmentation ability when processing biomedical images with multiple-shaped targets. At the same time, this model can significantly reduce the phenomenon of missed segmentation and missegmentation, improve segmentation accuracy, and make the segmentation results closer to the real situation.
- biomedical image segmentation,
- deep learning,
- dense atrous convolution,
- dense residual pooling,
- multi-scale features,
- attention mechanism
Citation: Hong'an Li, Man Liu, Jiangwen Fan, Qingfang Liu. Biomedical image segmentation algorithm based on dense atrous convolution[J]. Mathematical Biosciences and Engineering, 2024, 21(3): 4351-4369. doi: 10.3934/mbe.2024192

Related Papers:

Abstract

Biomedical images have complex tissue structures, and there are great differences between images of the same part of different individuals. Although deep learning methods have made some progress in automatic segmentation of biomedical images, the segmentation accuracy is relatively low for biomedical images with significant changes in segmentation targets, and there are also problems of missegmentation and missed segmentation. To address these challenges, we proposed a biomedical image segmentation method based on dense atrous convolution. First, we added a dense atrous convolution module (DAC) between the encoding and decoding paths of the U-Net network. This module was based on the inception structure and atrous convolution design, which can effectively capture multi-scale features of images. Second, we introduced a dense residual pooling module to detect multi-scale features in images by connecting residual pooling blocks of different sizes. Finally, in the decoding part of the network, we adopted an attention mechanism to suppress background interference by enhancing the weight of the target area. These modules work together to improve the accuracy and robustness of biomedical image segmentation. The experimental results showed that compared to mainstream segmentation networks, our segmentation model exhibited stronger segmentation ability when processing biomedical images with multiple-shaped targets. At the same time, this model can significantly reduce the phenomenon of missed segmentation and missegmentation, improve segmentation accuracy, and make the segmentation results closer to the real situation.

References

[1]	L. Kong, Q. Wang, Y. Bao, H. Li, A survey on medical image segmentation based on deep leaning, Radio Commun. Technol., 47 (2021), 121–130. https://doi.org/10.3969/j.issn.1003-3114.2021.02.001 doi: 10.3969/j.issn.1003-3114.2021.02.001
[2]	J. Chen, L. Li, Automatic segmentation of fuzzy areas in ultrasonic images based on edge detection, Autom. Instrument., 11 (2021), 19–22. https://doi.org/10.14016/j.cnki.1001-9227.2021.11.019 doi: 10.14016/j.cnki.1001-9227.2021.11.019
[3]	A. Aslam, E. Khan, M. M. S. Beg, Improved edge detection algorithm for brain tumor segmentation, Proced. Computer Sci., 58 (2015), 430–437. https://doi.org/10.1016/j.procs.2015.08.057 doi: 10.1016/j.procs.2015.08.057
[4]	M. Van Eijnatten, R. van Dijk, J. Dobbe, G. Streekstra, J. Koivisto, J. Wolff, CT image segmentation methods for bone used in medical additive manufacturing, Med. Eng. Phys., 51 (2018), 6–16. https://doi.org/10.1016/j.medengphy.2017.10.008 doi: 10.1016/j.medengphy.2017.10.008
[5]	J. Long, E. Shelhamer, T. Darrell, Fully convolutional networks for semantic segmentation, in Proceedings of the IEEE conference on computer vision and pattern recognition, (2015), 3431–3440. https://doi.org/10.1109/CVPR.2015.7298965
[6]	O. Ronneberger, P. Fischer, T. Brox, U-net: Convolutional networks for biomedical image segmentation, in International Conference on Medical image computing and computer assisted intervention, (2015), 234–241. https://doi.org/10.1007/978-3-319-24574-4_28
[7]	O. Oktay, J. Schlemper, L. L. Folgoc, M. Lee, M. Heinrich, K. Misawa, et al., Attention u-net: Learning where to look for the pancreas, arXiv preprint, (2018), arXiv: 1804.03999. https://doi.org/10.48550/arXiv.1804.03999
[8]	Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh, J. Liang, Unet++: A nested u-net architecture for medical image segmentation, Deep learning in medical image analysis and multimodal learning for clinical decision support, (2018) 3–11. https://doi.org/10.1007/978-3-030-00889-5_1 doi: 10.1007/978-3-030-00889-5_1
[9]	M. Z. Alom, M. Hasan, C. Yakopcic, T. M. Taha, V. K. Asari, Recurrent residual convolutional neural network based on u-net (r2u-net) for medical image segmentation, arXiv preprint, (2018), arXiv: 1802.06955. https://doi.org/10.48550/arXiv.1802.06955
[10]	G. Huang, S. Liu, L. Van der Maaten, K. Q. Weinberger, Condensenet: An efficient densenet using learned group convolutions, in Proceedings of the IEEE conference on computer vision and pattern recognition, (2018), 2752–2761. https://doi.org/10.1109/CVPR.2018.00291
[11]	F. Zhu, Z. Gao, C. Zhao, Z. Zhu, J. Tang, Y. Liu, et al., Semantic segmentation using deep learning to extract total extraocular muscles and optic nerve from orbital computed tomography images, Optik, 244 (2021), 167551. https://doi.org/10.1016/j.ijleo.2021.167551 doi: 10.1016/j.ijleo.2021.167551
[12]	H. Li, J. Fan, Q. Hua, X. Li, Z. Wen, M. Yang, Biomedical sensor image segmentation algorithm based on improved fully convolutional network, Measurement, 197 (2022), 111307. https://doi.org/10.1016/j.measurement.2022.111307 doi: 10.1016/j.measurement.2022.111307
[13]	X. Liu, D. Zhang, J. Yao, J. Tang, Transformer and convolutional based dual branch network for retinal vessel segmentation in OCTA images, Biomed. Signal Process. Control, 83 (2023). https://doi.org/10.1016/j.bspc.2023.104604 doi: 10.1016/j.bspc.2023.104604
[14]	N. Mu, Z. Lyu, M. Rezaeitaleshmahalleh, J. Tang, J. Jiang, An attention residual u-net with differential preprocessing and geometric postprocessing: Learning how to segment vasculature including intracranial aneurysms, Med. Image Anal., 84 (2023), 102697. https://doi.org/10.1016/j.media.2022.102697 doi: 10.1016/j.media.2022.102697
[15]	J. Qin, J. Wu, P. Yan, M. Li, R. Yuxi, X. Xiao, et al., FreeSeg: unified, universal and open-vocabulary image segmentation, inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2023), 19446–19455. https://doi.org/10.1109/CVPR52729.2023.01863
[16]	T. Ma, H. Zhao, X. Qin, A dehazing method for flight view images based on transformer and physical priori, Math. Biosci. Eng.*, 20 (2023), 20727–20747. http://dx.doi.org/10.3934/mbe.2023917 doi: 10.3934/mbe.2023917
[17]	A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, et al., Segment anything, arXiv preprint, (2023), arXiv: 2304.02643. https://doi.org/10.48550/arXiv.2304.02643
[18]	X. Zhao, W. Ding, Y. An, Y. Du, T. Yu, M. Li, et al., Fast Segment Anything, arXiv preprint, (2023), arXiv: 2306.12156. https://doi.org/10.48550/arXiv.2306.12156
[19]	C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, et al., Going deeper with convolutions, in Proceedings of the IEEE conference on computer vision and pattern recognition, (2015), 1–9. https://doi.org/10.1109/CVPR.2015.7298594
[20]	S. Ioffe, C. Szegedy, Batch normalization: accelerating deep network training by reducing internal covariate shift, in International conference on machine learning, (2015), 448–456. https://doi.org/10.48550/arXiv.1502.03167
[21]	C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, Z. Wojna, Rethinking the inception architecture for computer vision, in Proceedings of the IEEE conference on computer vision and pattern recognition, (2016), 2818–2826. https://doi.org/10.1109/CVPR.2016.308
[22]	C. Szegedy, S. Ioffe, V. Vanhoucke, A. Alemi, Inception-v4, inception-resnet and the impact of residual connections on learning, in Proceedings of the AAAI conference on artificial intelligence, 31 (2017). https://doi.org/10.48550/arXiv.1602.07261
[23]	T. Ma, C. Fu, J. Yang, J. Zhang, C. Shang, RF-Net: unsupervised low-light image enhancement based on retinex and exposure fusion, Comput. Mater. Continua, 77 (2023), 1103–1122. https://doi.org/10.32604/cmc.2023.042416 doi: 10.32604/cmc.2023.042416
[24]	C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, et al., Going deeper with convolutions, in Proceedings of the IEEE conference on computer vision and pattern recognition, (2015), 1–9. https://doi.org/10.1109/CVPR.2015.7298594
[25]	Y. Wei, H. Xiao, H. Shi, Z. Jie, J. Feng, T. S. Huang, Revisiting dilated convolution: A simple approach for weakly and semi supervised semantic segmentation, in Proceedings of the IEEE conference on computer vision and pattern recognition, (2018), 7268–7277. https://doi.org/10.1109/CVPR.2018.00759
[26]	D. M. Vo, S. W. Lee, Semantic image segmentation using fully convolutional neural networks with multi-scale images and multi-scale dilated convolutions, Multimedia Tools Appl., 77 (2018), 18689–18707. https://doi.org/10.1007/s11042-018-5653-x doi: 10.1007/s11042-018-5653-x
[27]	H. Li, Q. Zheng, W. Yan, R. Tao, X. Qi, Z. Wen, Image super-resolution reconstruction for secure data transmission in Internet of Things environment, Math. Biosci. Eng., 18 (2021), 6652–6672. https://doi.org/10.3934/mbe.2021330 doi: 10.3934/mbe.2021330
[28]	H. Cheng, J. Lu, M. Luo, W. Liu, K. Zhang, PTANet: Triple attention network for point cloud semantic segmentation, Eng. Appl. Artif. Intell., 102 (2021), 104239. https://doi.org/10.1016/j.engappai.2021.104239 doi: 10.1016/j.engappai.2021.104239
[29]	Z. Gu, J. Cheng, H. Fu, K. Zhou, H. Hao, Y. Zhao, et al., Ce-net: Context encoder network for 2d medical image segmentation, IEEE Transact. Med. Imag., 38 (2019), 2281–2292. https://doi.org/10.1109/TMI.2019.2903562 doi: 10.1109/TMI.2019.2903562

Reader Comments

Your name:*

Email:*
© 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)