Automatic segmentation algorithm of pancreatic CT images based on 2-3D mixed convolutional neural network

Minquan NING; Jing WEN; Yi HUANG; Yi WANG

doi:10.16016/j.1000-5404.202103046

您当前的位置：

首页 >

文章列表页 >

Automatic segmentation algorithm of pancreatic CT images based on 2-3D mixed convolutional neural network

Monographic Reports: Digital medicine and artificial intelligence | 更新时间：2026-01-09

- Automatic segmentation algorithm of pancreatic CT images based on 2-3D mixed convolutional neural network
- Journal of Third Military Medical University Vol. 43, Issue 18, Pages: 1729-1734(2021)
- 作者机构：
  
  400044 重庆，重庆大学计算机学院
- 作者简介：
- 基金信息：
  
  the General Program of National Natural Science Foundations of China(61672120);the Fundamental Research Funds for Central Universities(2020CDCGJSJ043);the Fundamental Research Funds for Central Universities(2020CDCGSL054);the Fundamental Research Funds for Central Universities(2019CDYGZD004)
- DOI：10.16016/j.1000-5404.202103046
  CLC： R319;R322.491;R814.42
- Received：04 March 2021，
  
  Revised：2021-05-17，
  
  Published：30 September 2021
- 稿件说明：
移动端阅览
Minquan NING, Jing WEN, Yi HUANG, et al. Automatic segmentation algorithm of pancreatic CT images based on 2-3D mixed convolutional neural network[J]. Journal of Third Military Medical University, 2021, 43(18): 1729-1734.
DOI：

Minquan NING, Jing WEN, Yi HUANG, et al. Automatic segmentation algorithm of pancreatic CT images based on 2-3D mixed convolutional neural network[J]. Journal of Third Military Medical University, 2021, 43(18): 1729-1734. DOI： 10.16016/j.1000-5404.202103046.

摘要

目的

探讨CT图像中胰腺组织自动分割的深度学习算法。

方法

以2D胰腺分割网络和3D胰腺分割网络为基础，利用2D分割网络提取的判别性语义信息，3D分割网络提取出三维空间信息，最后将判别性的语义信息和三维空间信息进行融合实现胰腺分割。

结果

本研究在NIH数据集上的进行了验证，平均DICE系数达到83.2%(最大值90.65%，最小值67.04%)。分别超出了2D基准方法和3D基准方法0.77%和1.38%。其胰腺分割结果细节准确，边缘平滑。2-3D方法分割出的胰腺组织相比于2D方法和3D方法分割出的胰腺，在轮廓形态上与对应的手工标注图吻合度更高，更准确。相比于2D方法，2-3D方法有效综合了三维空间信息，对胰腺边缘位置进行了补足，避免了在胰腺边缘出现的漏分割情况。相比于3D方法，2-3D方法有效地利用了2D方法提供的语义信息，规避了在胰腺边缘出现的错误分割的情况。

结论

该方法融合了2D分割网络和3D分割网络的优点，能够有效地应用于对胰腺的准确分割。

Abstract

Objective

To investigate the deep learning algorithm for automatic segmentation of pancreatic tissue from CT images.

Methods

Based on both 2D and 3D pancreatic segmentation networks

which extracted discriminative semantic information and three-dimensional spatial information from 82 CT images of pancreas samples (taken from NIH database) respectively

a novel hybrid network was designed for automatic segmentation of pancreatic tissues.

Results

The proposed method was validated using NIH datasets

and the Results of pancreas segmentation demonstrated accurately in detail and smooth in edge. The average Dice coefficient reached 83.2%

which exceeded the 2D and 3D benchmark Methods by 0.77% and 1.38%

respectively.

Conclusion

This method combines both advantages of 2D and 3D segmentation networks

and thus can be effectively applied to the accurate segmentation of pancreas

providing a quantitative reference for subsequent clinical diagnosis and treatment.

关键词

Keywords

references

CHAN-SENG-YUE M, KIM J C, WILSON G W, et al . Transcription phenotypes of pancreatic cancer are driven by genomic events during tumor evolution[J ] . Nat Genet, 2020, 52(2): 231-240. DOI: 10.1038/s41588-019-0566-9.

何铭, 薛华丹, 金征宇. 影像组学在胰腺癌的应用及研究进展[J]. 中华放射学杂志, 2020, 54(10): 1029-1032. DOI: 10.3760/cma.j.cn112149-20190903-00738.

HE M, XUE H D, JIN Z Y. Radiomics in pancreatic cancer: the state of art[J]. Chin J Radiol, 2020, 54(10): 1029-1032. DOI: 10.3760/cma.j.cn112149-20190903-00738.

李仓, 吴晓康, 厉学民, 等. 三维可视化技术结合3D腹腔镜在胰腺中段切除术中的临床应用[J]. 中华胰腺病杂志, 2020(6): 434-437. DOI: 10.3760/cma.j.cn115667-20200610-00090.

LI C, WU X K, LI X M, et al . Clinical application of 3D visualization technology combined with 3D laparoscopy in the resection of middle segment pancreatectomy[J ] . Chin J Pancreatol, 2020(6): 434-437. DOI: 10.3760/cma.j.cn115667-20200610-00090

ZHOU Y Y, XIE L X, SHEN W, et al . A fixed-point model for pancreas segmentation in abdominal CT scans[M ] //Medical Image Computing and Computer Assisted Intervention-MICCAI 2017. Cham: Springer International Publishing, 2017: 693-701. DOI: 10.1007/978-3-319-66182-7_79

ROTHH R, SHEN C, ODA H, et al . Deep learning and its application to medical image segmentation[J ] . Med Imaging Technol, 2018, 36. DOI: 10.11409/mit.36.63

RONNEBERGER O, FISCHER P, BROX T. U-Net: Convolutional Networks for Biomedical Image Segmentation[C]. MICCAI, 2015, 9351. 234-241. DOI: 10.1007/978-3-319-24574-4_28.

OKTAYO, SCHLEMPER J, FOLGOC L L, et al . Attention U-net: learning where to look for the pancreas[EB/OL ] . 2018.

ZHOUX R, ITO T, TAKAYAMA R, et al . Three-dimensional CT image segmentation by combining 2D fully convolutional network with 3D majority voting[M ] //Deep Learning and Data Labeling for Medical Applications. Cham: Springer International Publishing, 2016: 111-120. DOI: 10.1007/978-3-319-46976-8_12.

ÇIÇEKÖ, ABDULKADIRA, LIENKAMP S S, et al . 3D U-net: learning dense volumetric segmentation from sparse annotation[C ] //Med Image Comput Comput Assist Interv MICCAI, 2016: 424-432. DOI: 10.1007/978-3-319-46723-8_49.

MILLETARIF, NAVAB N, AHMADI S A. V-net: fully convolutional neural networks for volumetric medical image segmentation[C]//2016 Fourth International Conference on 3D Vision (3DV). October 25-28, 2016, Stanford, CA, USA. IEEE, 2016: 565-571. DOI: 10.1109/3DV.2016.79.

ROTHH R, ODA H, ZHOU X R, et al . An application of cascaded 3D fully convolutional networks for medical image segmentation[J ] . Comput Med Imaging Graph, 2018, 66: 90-99. DOI: 10.1016/j.compmedimag.2018.03.001.

NWE T L, MIN O Z, GOPALAKRISHNAN S, et al . Improving 3D brain tumor segmentation with predict-refine mechanism using saliency and feature maps[C ] //2020 IEEE International Conference on Image Processing (ICIP). October 25-28, 2020, Abu Dhabi, United Arab Emirates. IEEE, 2020: 2671-2675. DOI: 10.1109/ICIP40778.2020.9190806.

Roth H R, Lu L, Farag A, et al . DeepOrgan: Multi-level Deep Convolutional Networks for Automated Pancreas Segmentation[C ] //International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2015: 556-564. DOI: 10.1007/978-3-319-24553-9_68.

HE K M, ZHANG X Y, REN S Q, et al . Deep residual learning for image recognition[C ] //2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). June 27-30, 2016, Las Vegas, USA. IEEE, 2016: 770-778. DOI: 10.1109/CVPR.2016.90.

SHELHAMERE, LONG J, DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Trans Pattern Anal Mach Intell, 2017, 39(4): 640-651. DOI: 10.1109/TPAMI.2016.2572683.

ROTH H R, ODA H, ZHOU X R, et al . An application of cascaded 3D fully convolutional networks for medical image segment ation[J ] . Comput Med Imaging Graph, 2018, 66: 90-99. DOI: 10.1016/j.compmedimag.2018.03.001.

郑彩侠, 张同舟, 孙长江, 等. 图像分割方法在医学领域中的应用[J]. 中国医疗设备, 2018, 33(6): 1-5, 11. DOI: 10.3969/j.issn.1674-1633.2018.06.001.

ZHENG C X, ZHANG T Z, SUN C J, et al . Application of image segmentation methods in the medical field[J ] . China Med Devices, 2018, 33(6): 1-5, 11. DOI: 10.3969/j.issn.1674-1633.2018.06.001.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

No data

Related Author

No data

Related Institution

No data

⁰