基于改进UNet++模型的染色体图像分割研究
Research on Chromosome Image Segmentation Based on Improved UNet++ Model
DOI: 10.12677/MOS.2024.132135, PDF,   
作者: 马秋怡, 张学典：上海理工大学光电信息与计算机工程学院，上海
关键词: UNet++；染色体分割；CBAM注意力机制；模型轻量化；UNet++； Chromosome Segmentation； CBAM Attention Mechanism； Lightweight Model
摘要: 关于人类23对染色体的遗传学研究已经广泛应用于多个领域，尤其针对于染色体数目形态与染色体疾病的研究。研究者们采用了各类深度学习模型以提升重叠染色体分割的精确度。本文通过集合不同深度的UNets、同时嵌套稠密卷积块来改进跳跃连接，使用深度学习方法以保持不同大小的研究对象都有良好的分割性能。综上，本文提出MCBA-UNet模型，在改进后的UNet++模型上引入了MobileNet轻量化模型，解决了因数据过多导致计算机性能下降的问题，再使用CBAM注意力机制校准其分割准确性。实验结果证明，MCBA-UNet对于重叠染色体的分割准确率可以达到98.73%。
Abstract: The genetic research on human 23 pairs of chromosomes has been widely applied in multiple fields, especially in the study of chromosome number, morphology, and chromosomal diseases. Research-ers have adopted various deep learning models to improve the accuracy of overlapping chromo-some segmentation. This article improves skip connections by combining UNets of different depths and simultaneously nesting dense convolutional blocks. Deep learning methods are used to main-tain good segmentation performance for research objects of different sizes. In summary, this article proposes the MCBA UNet model, which introduces the MobileNet lightweight model on the im-proved UNet++ model to solve the problem of computer performance degradation caused by exces-sive data. Then, the CBAM attention mechanism is used to calibrate its segmentation accuracy. The experimental results show that the segmentation accuracy of MCBA-UNet for overlapping chromo-somes can reach 98.73%.
文章引用：马秋怡, 张学典. 基于改进UNet++模型的染色体图像分割研究[J]. 建模与仿真, 2024, 13(2): 1434-1443. https://doi.org/10.12677/MOS.2024.132135

参考文献

[1] Zhu, L.L., Han, L. and Du, H. (2020) Etc Research on Multi Active Contour Cell Segmentation Method Based on U-Net Net-work. Infrared and Laser Engineering, 49, 145-153.
[2] You, Q.J. (2020) The Application of Machine Learning In Chromo-some and Fundus Image Analysis. Master’s Thesis, Nanjing University of Aeronautics and Astronautics, Nanjing.
[3] Xie, Y.Z., Yan, S.J., Wei, G.F., et al. (2022) Breast Mass Segmentation Based on U-Net++ and Adversarial Learning Networks. Laser & Optoelectronics Progress, 59, Article ID: 1617002.
[4] Ronneberger, O., Fischer, P. and Thomas, B. (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. In: Navab, N., Hornegger, J., Wells, W. and Frangi, A., Eds., MICCAI 2015: Medical Image Computing and Computer-Assisted Intervention—MICCAI 2015, Springer, Cham, 234-241. [Google Scholar] [CrossRef] 
[5] Pijackova, K., Gotthans, T. and Gotthans, J. (2022) Deep Learning Pipeline for Chromosome Segmentation. 2022 32nd International Conference Radioelektronika (RADIOELEKTRONIKA), Kosice, 21-22 April 2022, 1-5. [Google Scholar] [CrossRef] 
[6] Saleh, H.M., Saad, N.H. and Mat Isa, N.A. (2019) Overlapping Chromosome Segmentation Using U-Net: Convolutional Networks with Test Time Augmentation. Proce-dia Computer Science, 159, 524-533. [Google Scholar] [CrossRef] 
[7] Cao, X., Lan, F., Liu, C.M., Lam, T.W. and Luo, R. (2020) ChromSeg: Two-Stage Framework for Overlapping Chromosome Segmentation and Reconstruction. 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Seoul, 16-19 December 2020, 2335-2342. [Google Scholar] [CrossRef] 
[8] Zhao, X., Yuan, Y.T., Song, M.D., et al. (2019) Use of Un-manned Aerial Vehicle Imagery and Deep Learning UNet to Extract Rice Lodging. Sensors, 19, Article 3859. [Google Scholar] [CrossRef] [PubMed]
[9] Sae-Lim, W., Wettayaprasit, W. and Aiyarak, P. (2019) Convolutional Neural Networks Using MobileNet for Skin Lesion Classification. Proceedings of the 16th International Joint Conference on Comput-er Science and Software Engineering, Chonburi, 10-12 July 2019, 242-247. [Google Scholar] [CrossRef] 
[10] Wang, W., Hu, Y., Zou, T., Liu, H., Wang, J. and Wang, X. (2020) A New Image Classification Approach via Improved MobileNet Models with Local Receptive Field Expansion In Shallow Layers. Computational Intelligence and Neuroscience, 2020, Article ID: 8817849. [Google Scholar] [CrossRef] [PubMed]
[11] Liu, X., Jia, Z., Hou, X. and Fu, M. (2019) Real-Time Marine Animal Images Classification by Embedded System Based on Mo-bilenet and Transfer Learning. Proceedings of the OCEANS 2019, Marseille, 17-20 June 2019, 1-5. [Google Scholar] [CrossRef] 
[12] Bi, C., Wang, J. and Duan, Y. (2020) MobileNet Based Apple Leaf Diseases Identification. Mobile Networks and Applications, 27, 172-180. [Google Scholar] [CrossRef] 
[13] Woo, S., Park, J., Lee, J.Y. and Kweon, I.S. (2018) CBAM: Convolu-tional Block Attention Module. In: Ferrari, V., Hebert, M., Sminchisescu, C. and Weiss, Y., Eds., Computer Vision—ECCV 2018, Springer, Cham, 3-19. [Google Scholar] [CrossRef] 
[14] 王志勇, 刘溢凡, 张杰, 等. 基于“AM-GoogLeNet+BP”联合数据驱动的混凝土细观模型压缩应力-应变曲线预测[J]. 力学学报, 2023, 55(4): 925-938.
[15] Zhou, Z., Siddiquee, M.M.R., Tajbakhsh, N. and Liang, J.M. (2020) UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Seg-mentation. IEEE Transactions on Medical Imaging, 39, 1856-1867. [Google Scholar] [CrossRef] 
[16] Zhou, Z., Rahman Siddiquee, M.M., Tajbakhsh, N., et al. (2018) Unet++: A Nested U-Net Architecture for Medical Image Segmentation. In: Stoyanov, D., et al. Eds., Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, Springer, Cham, 3-11. [Google Scholar] [CrossRef] [PubMed]
[17] Jeanpat (2016) Overlapping Chromosomes.
https://www.kaggle.com/datasets/jeanpat/overlapping-chromosomes
[18] Saleh, H.M., Saad, N.H. and Isa, N.A.M. (2019) Overlapping Chromosome Segmentation Using U-Net: Convolutional Networks with Test Time Augmentation. Procedia Computer Science, 159, 524-533. [Google Scholar] [CrossRef] 
[19] Huang, H., Lv, R.F., Tao, J.l., et al. (2021) Etc CT Image Lung Nodule Segmentation Algorithm Based on Improved U-Net++. Acta Photonica Sinica, 50, 73-83. [Google Scholar] [CrossRef] 
[20] Lin, C.C., Zhao, G.S., Yin, A.H., et al. (2020) AS-PANet: Im-proved Path Enhancement Network for Overlapping Chromosome Instance Segmentation. Chinese Journal of Image and Graphics, 25, 2271-2280.
[21] Cao, X., Lan, F., Liu, C, M., et al. (2020) ChromSeg: Two-Stage Framework for Overlapping Chromosome Segmentation and Reconstruction. 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Seoul, 16-19 December 2020, 2335-2342. [Google Scholar] [CrossRef] 
[22] Zhang, J. (2023) Image Style Transfer Based on DeepLabV3+ Semantic Segmentation. Highlights in Science, Engineering and Technolo-gy, 39, 1203-1213. [Google Scholar] [CrossRef] 
[23] Badrinarayanan, V., Kendall, A. AND Cipolla, R. (2017) SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39, 2481-2495. [Google Scholar] [CrossRef] 
[24] Kluvanec, D., Phillips Thomas, B., Mccaffrey Kenneth, J.W. and Al Moubayed, N. (2022) Using Orientation to Distinguish Overlapping Chromosomes. In: Pimenidis, E., Angelov, P., Jayne, C., Papaleonidas, A. and Aydin, M., Eds., Artificial Neural Networks and Machine Learning—ICANN 2022, Springer, Cham, 391-403.

为你推荐