一种新型的肾实质分割网络
A New Type of Renal Parenchymal Segmentation Network
摘要: 肾是人体内十分重要的一个组织器官，因此肾相关的各种病最近几年引起了极大的关注。在这之中，肾实质就是其中的最常见的一种肾病之一。到目前为止，关于肾实质病变的诊断主要依赖于临床医生的标注，从而人工进行判断。这样的方式需要很大的人工与时间成本，因此，亟需一种新的方法提升肾实质诊断的效率和精度。本文主要针对小儿肾图，建立了相关的儿童肾实质数据集。并且，根据小儿肾图数据集的特征，我们提出了一种新的分割方法。本方法的核心是细化跳跃连接模块(RSC模块)与transform架构。本文提出的网络不仅提升了分割的精度，改善了由于分辨率降低从而导致的感受野下降的问题，而且大大减少了人工标注的时间，提高了诊断的精度。本文的代码是基于pytorch框架进行的编程，在小儿肾图数据集进行的实验，此外，将本文提出的网络与经典的FCN、SegNet、U-Net和Deeplab-V3+做了对比实验。结果显示本文提出的方法在precision、dice_coeff、recall三种评价指标上，(对比其网络在这三种指标上最优的结果)分别提升了2.547%、4.992%、2.498%，其效果也得到了专业医生的认可。
Abstract: The kidney is a very important tissue and organ in the human body, so various diseases related to the kidney have attracted great attention in recent years. Among them, the renal parenchyma is one of the most common kidney diseases. Until now, the diagnosis of renal parenchymal lesions has relied mainly on the clinician’s annotations to make manual judgments. This method requires a lot of labor and time costs, so there is an urgent need for a new method to improve the efficiency and accuracy of renal parenchymal diagnosis. In this paper, the relevant pediatric renal parenchymal dataset is established for pediatric renal maps. And, based on the characteristics of the pediatric kidney map dataset, we propose a new segmentation method. The core of this method is to refine the jump connection module (RSC module) and transform architecture. The network proposed in this paper not only improves the accuracy of segmentation and improves the problem of declining the sensing field due to the reduction of resolution, but also greatly reduces the time of manual labeling and improves the accuracy of diagnosis. The code in this article is based on the pytorch framework, experiments performed on pediatric kidney map datasets, and in addition, the network proposed in this paper is compared with the classic FCN, SegNet, U-Net, and Deeplab-V3+. The results show that the proposed method is based on three evaluation indicators: precision, dice_coeff and recall. (Comparing the optimal results of its network on these three indicators) it is increased by 2.547%, 4.992%, and 2.498% respectively, and its effect was also recognized by professional doctors.
文章引用：张容祥. 一种新型的肾实质分割网络[J]. 理论数学, 2022, 12(10): 1661-1668. https://doi.org/10.12677/PM.2022.1210180

参考文献

[1] Chan, H.P., Doi, K., Galhotra, S., Vyborny, C.J., MacMahon, H. and Jokich, P.M. (1987) Image Feature Analysis and Computer-Aided Diagnosis in Digital Radiography. I. Automated Detection of Microcalcifications in Mammography. Medical Physics, 14, 538-548. [Google Scholar] [CrossRef] [PubMed]
[2] Van Ginneken, B., Romeny, B.M.T.H. and Viergever, M.A. (2001) Computer-Aided Diagnosis in Chest Radiography: A Survey. IEEE Transactions on Medical Imaging, 20, 1228-1241. [Google Scholar] [CrossRef] [PubMed]
[3] Jha, D., Smedsrud, P.H., Riegler, M.A., Halvorsen, P., de Lange, T., Johansen, D., et al. (2020) Kvasir-Seg: A Segmented Polyp Dataset. International Con-ference on Multimedia Modeling, Daejeon, 5-8 January 2020, 451-462. [Google Scholar] [CrossRef] 
[4] Ronneberger, O., Fischer, P. and Brox, T. (2015) U-net: Convolutional Networks for Biomedical Image Segmentation. 18th International Conference on Medical Image Com-puting and Computer-Assisted Intervention, Munich, 5-9 October 2015, 234-241. [Google Scholar] [CrossRef] 
[5] Long, J., Shelhamer, E. and Darrell, T. (2015) Fully Convo-lutional Networks for Semantic Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, 7-12 June 2015, 3431-3440. [Google Scholar] [CrossRef] 
[6] Chen, L.C., Zhu, Y., Papandreou, G., Schroff, F. and Adam, H. (2018) Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. Proceedings of the European Conference on Computer Vision (ECCV) 2018, Munich, 8-14 September 2018, 833-851. [Google Scholar] [CrossRef] 
[7] 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] 
[8] Hu, J., Shen, L., Albanie, S., Sun, G. and Wu, E. (2020) Squeeze-and-Excitation Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42, 2011-2023. [Google Scholar] [CrossRef] 
[9] Girshick, R., Donahue, J., Darrell, T. and Malik, J. (2014) Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, 23-28 June 2014, 580-587. [Google Scholar] [CrossRef] 
[10] Girshick, R. (2015) Fast R-CNN. Proceedings of the IEEE Confer-ence on Computer Vision and Pattern Recognition, Santiago, 7-13 December 2015, 1440-1448. [Google Scholar] [CrossRef] 
[11] Redmon, J., Divvala, S., Girshick, R. and Farhadi, A. (2016) You Only Look Once: Unified, Real-Time Object Detection. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, 27-30 June 2016, 779-788. [Google Scholar] [CrossRef] 
[12] Redmon, J. and Farhadi, A. (2017) YOLO9000: Better, Faster, Stronger. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition, Hawaii, 21-26 July 2017, 6517-6525. [Google Scholar] [CrossRef] 
[13] Redmon, J. and Farhad, A. (2018) YOLOv3: An Incremental Im-provement. arXiv e-prints, arXiv:1804.02767.
https://arxiv.org/abs/1804.02767
[14] He, K., Gkioxari, G., Dollár, P. and Girshick, R. (2017) Mask R-CNN. Proceedings of the IEEE Conference on Computer Vision, Venice, 22-29 October 2017, 2980-2988. [Google Scholar] [CrossRef] 
[15] Paszke, A., Gross, S., Massa, F., et al. (2019) Pytorch: An Impera-tive Style, High-Performance Deep Learning Library. 36th Annual Conference on Neural Information Processing Sys-tems, 8-14 December 2019, 8026-8037.

为你推荐