多分支压缩和激励网络的蕈样肉芽肿细胞分类

doi:10.13190/j.jbupt.2021-183

北京邮电大学学报 ›› 2022, Vol. 45 ›› Issue (4): 31-36.doi: 10.13190/j.jbupt.2021-183

多分支压缩和激励网络的蕈样肉芽肿细胞分类

王俊洁¹, 徐聪聪², 赵增瑞¹, 徐军¹, 姜祎群²

1. 南京信息工程大学自动化学院, 南京 210044;
2. 中国医学科学院北京协和医学院皮肤病研究所, 南京 210042

收稿日期:2021-08-28 出版日期:2022-08-28 发布日期:2022-09-03
通讯作者: 徐军(1973—),男,教授,博士生导师,邮箱:xujung@gmail.com。 E-mail:xujung@gmail.com
作者简介:王俊洁(1996—),女,硕士生。
基金资助:
国家自然科学基金项目（U1809205，62171230，61771249，81871352）

Classification of Mycosis Fungoides Cells Based on Multi Branch Squeeze and Excitation Network

WANG Junjie¹, XU Congcong², ZHAO Zengrui¹, XU Jun¹, JIANG Yiqun²

1. School of Automation, Nanjing University of Information Science and Technology, Nanjing 210044, China;
2. Department of Pathology, Hospital of Dermatology, Chinese Academy of Medical Science and Peking Union Medical College, Nanjing 210042, China

Received:2021-08-28 Online:2022-08-28 Published:2022-09-03

摘要/Abstract

摘要： 基于中国医学科学院皮肤病研究所收集的77张早、中期皮肤蕈样肉芽肿全扫描切片，构建了多分支压缩和激励网络模型，实现了皮肤蕈样肉芽肿的淋巴样细胞与上皮细胞的分类。模型分为编码和解码2个阶段，编码阶段对应一个分支；解码阶段有3个分支，对应一个主任务和2个辅助任务。主任务分支输出细胞分类的结果，辅助分支I输出细胞与背景，辅助分支II输出水平垂直边界图谱。在训练阶段，从切片中选取576张图像块，由专业病理科医生进行标记，其中将464张图像块用于训练，将112张图像块用于验证，最后在全扫描切片上进行测试。模型的细胞分割准确率为0.943，F1值为0.728，细胞的分类平均准确率为0.943。实验结果表明，所提出的模型能够实现皮肤蕈样肉芽肿淋巴样细胞和上皮细胞的识别与分类，为皮肤蕈样肉芽肿的计算机辅助诊断奠定了重要基础。

关键词: 蕈样肉芽肿, 多分支压缩与激励网络, 淋巴样细胞, 水平垂直边界图谱

Abstract: To study the different cell components of mycosis fungoides, a multi branch squeeze and excitation network model is constructed based on 77 whole slide images of early and middle stage mycosis fungoides, and the classification of lymphocytes and epithelial cells of mycosis fungoides is realized. The network is divided into two stages:encoding and decoding. The encoding stage corresponds to one branch, and the decoding stage has three branches, corresponding to one main task and two auxiliary tasks. The main task branch outputs the results of cell classification, the auxiliary branch I outputs the cells and background, and the auxiliary branch II outputs the horizontal and vertical boundary map. In the training stage, 576 image blocks were selected from the slices and marked by professional pathologists, including 464 for training and 112 for verification. Finally, they are tested on the whole slide images. The cell segmentation accuracy and F1 score of the model are 0.943 and 0.728, respectively. The average accuracy of classification is 0.943. The experimental results show that the proposed model can recognize and classify lymphocytes and epithelial cells in mycosis fungoides, which lays an important foundation for computer-aided diagnosis of cutaneous mycosis fungoides.

Key words: mycosis fungoides, multi branch squeeze and excitation network, lymphocytes cells, horizontal and vertical map

中图分类号:

TP391.4

王俊洁, 徐聪聪, 赵增瑞, 徐军, 姜祎群. 多分支压缩和激励网络的蕈样肉芽肿细胞分类[J]. 北京邮电大学学报, 2022, 45(4): 31-36.

WANG Junjie, XU Congcong, ZHAO Zengrui, XU Jun, JIANG Yiqun. Classification of Mycosis Fungoides Cells Based on Multi Branch Squeeze and Excitation Network[J]. Journal of Beijing University of Posts and Telecommunications, 2022, 45(4): 31-36.

参考文献

[1] 林春玲, 赵延明, 牛云彤. 68例皮肤恶性肿瘤与癌前病变的临床分析[J]. 实用癌症杂志, 2017, 32(6):1031-1033. LIN C L, ZHAO Y M, NIU Y T. Clinical analysis of 68 cases of skin malignant tumor and precancerous lesion[J]. The Practical Journal of Cancer, 2017, 32(6):1031-1033.
[2] 张层层, 吴娟, 徐令祥. 蕈样肉芽肿[J]. 世界最新医学信息文摘, 2015(29):75. ZHANG C C, WU J, XU L X. Granuloma fungoides[J]. World Latest Medicine Information, 2015, 15(29):75.
[3] VONDERHEID E C, KADIN M E, TELANG G H. Papular mycosis fungoides:six new cases and association with chronic lymphocytic leukemia[J]. World Journal of Dermatology, 2016, 5(4):136-143.
[4] SCHMITZ L, KANITAKIS J. Histological classification of cutaneous squamous cell carcinomas with different severity[J]. Journal of the European Academy of Dermatology and Venereology:JEADV, 2019, 33(Sup 8):11-15.
[5] D'ACUNTO M, MARTINELLI M, MORONI D. From human mesenchymal stromal cells to osteosarcoma cells classification by deep learning[J]. Journal of Intelligent and Fuzzy Systems, 2019, 37(6):7199-7206.
[6] YUAN Y Y, FAILMEZGER H, RUEDA O M, et al. Quantitative image analysis of cellular heterogeneity in breast tumors complements genomic profiling[J]. Science Translational Medicine, 2012, 4(157):143-157.
[7] XU J, XIANG L, LIU Q S, et al. Stacked sparse autoencoder (SSAE) for nuclei detection on breast cancer histopathology images[J]. IEEE Transactions on Medical Imaging, 2016, 35(1):119-130.
[8] TOMCZAK A, ILIC S, MARQUARDT G, et al. Multi-task multi-domain learning for digital staining and classification of leukocytes[J]. IEEE Transactions on Medical Imaging, 2021, 40(10):2897-2910.
[9] GRAHAM S, VU Q D, RAZA S E A, et al. Hover-net:simultaneous segmentation and classification of nuclei in multi-tissue histology images[J]. Medical Image Analysis, 2019, 58:101563-101566.
[10] WANG P, HU X L, LI Y M, et al. Automatic cell nuclei segmentation and classification of breast cancer histopathology images[J]. Signal Processing, 2016, 122:1-13.
[11] XIANG H, YAN B, CAI Q, et al. An edge detection algorithm based-on Sobel operator for images captured by binocular microscope[C]//2011 International Conference on Electrical and Control Engineering. Piscataway, NJ:IEEE Press, 2011:980-982.
[12] HU J, SHEN L, ALBANIE S, et al. Squeeze-and-excitation networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8):2011-2023.
[13] ZHANG Y, LIU S J, LI C L, et al. Rethinking the dice loss for deep learning lesion segmentation in medical images[J]. Journal of Shanghai Jiaotong University (Science), 2021, 26(1):93-102.
[14] LEI Y, TIAN S B, HE X X, et al. Ultrasound prostate segmentation based on multidirectional deeply supervised V-Net[J]. Medical Physics, 2019, 46(7):3194-3206.
[15] LIN K W E, BALAMURALI B T, KOH E, et al. Singing voice separation using a deep convolutional neural network trained by ideal binary mask and cross entropy[J]. Neural Computing and Applications, 2020, 32(4):1037-1050.
[16] POUDEL P, BAE S H, JANG B. Comparison of different deep learning optimizers for modeling photovoltaic power[J]. Journal of the Chosun Natural Science, 2018, 11(4):204-208.
[17] LUO Y L. Multi-feature data mining for CT image recognition[J/OL]. Concurrency and Computation Practice and Experience, 2018, 32(3):e4885[2022-01-08]. https://www.zhangqiaokeyan.com/academic-journal-foreign_other_thesis/0204113718331.html.
[18] RONNEBERGER O, FISCHER P, BROX T. U-net:convolutional networks for biomedical image segmentation[M]//Lecture Notes in Computer Science. Cham:Springer International Publishing, 2015:234-241.

多分支压缩和激励网络的蕈样肉芽肿细胞分类

Classification of Mycosis Fungoides Cells Based on Multi Branch Squeeze and Excitation Network

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价