[1]贾欣齐,李 睿,张志成,等.DenseNet-GRU:直肠癌 CT 影像分类的深度神经网络模型[J].计算机技术与发展,2021,31(03):111-114.[doi:10. 3969 / j. issn. 1673-629X. 2021. 03. 019]
　JIA Xin-qi,LI Rui,ZHANG Zhi-cheng,et al.DenseNet-GRU:A Deep Neural Network Model for CT Image Classification of Rectal Cancer[J].,2021,31(03):111-114.[doi:10. 3969 / j. issn. 1673-629X. 2021. 03. 019]
点击复制

DenseNet-GRU:直肠癌 CT 影像分类的深度神经网络模型()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
31
期数:
2021年03期
页码:
111-114
栏目:
图形与图像
出版日期:
2021-03-10

文章信息/Info

Title:
DenseNet-GRU:A Deep Neural Network Model for CT Image Classification of Rectal Cancer
文章编号:
1673-629X(2021)03-0111-04
作者:
贾欣齐; 李 睿; 张志成; 王 阳; 吕 品
上海电机学院 电子信息学院,上海 201306
Author(s):
JIA Xin-qi; LI Rui; ZHANG Zhi-cheng; WANG Yang; LYU Pin
School of Electronics and Information,Shanghai Dianji University,Shanghai 201306,China
关键词:
DenseNet; 门控循环单元; 深度神经网络; 影像特征; 影像分类
Keywords:
DenseNet; gated recurrent unit; deep neural network; image feature; image classification
分类号:
TP391. 41
DOI:
10. 3969 / j. issn. 1673-629X. 2021. 03. 019
摘要:
DenseNet 是一种广泛用于影像分类的卷积神经网络,但它不具备记忆功能,无法反映卷积操作后不同特征映射之间的关联关系。 若将其直接应用于判断直肠癌是否发生淋巴结转移,则无法比较直肠癌 CT 影像特征在深度神经网络映射过程中的变化。 基于此,提出了一种新颖的深度神经网络模型 DenseNet-GRU(gated recurrent unit),其核心是利用 GRU获取 DenseNet 提取的不同影像特征之间的关联关系,进而获得不同图像之间相同像素区域的特征变化情况,最终判断直肠癌患者的淋巴结是否存在转移。 以包含 107 个患者 DCM 格式的腹部横断位动脉期和门脉期两种增强 CT 影像为实验数据集,采用数据增强和阈值分割方法对数据进行预处理,DenseNet-GRU 模型在 F-score 上的分类精度达到了 65% 以上,对临床辅助诊断具有重要的现实意义。
Abstract:
DenseNet is a convolutional neural network widely used in image classification,but it has no memory function and cannot reflect the correlation between different feature maps after convolutions. If it is directly used to judge whether there is lymph node metastasis in rectal cancer,it is impossible to compare the changes of rectal cancer CT image features in the process of feature map of deep neural networks. To resolve this problem,a novel deep neural network model DenseNet-GRU (gated recurrent unit) is proposed.The core? ? ? ?of DenseNet-GRU is to use GRU to obtain the correlation between different image features extracted by DenseNet. The feature changes of the same pixel area between different images can be captured by the correlation,and finally it is possible to judge whether there is lymph node metastasis in rectal cancer patients. The experimental dataset contains two kinds of arterial and portal phases of abdominal transection with DCM file format from 107 patients. The raw images are preprocessed by the methods of data enhancement and threshold segmentation. The classification accuracy of the proposed model on the F-score reaches beyond 65% ,which indicates that Densenet-GRU is effective and feasible for judging whether rectal cancers have lymph node metastasis and important for clinical auxiliary diagnosis.

相似文献/References:

[1]吴俊清,倪建成,魏媛媛.语音情感识别中面向小数据集的 CGRU 方法[J].计算机技术与发展,2020,30(12):77.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 014]
　WU Jun-qing,NI Jian-cheng,WEI Yuan-yuan.CGRU Method for Small Datasets in Speech Emotion Recognition[J].,2020,30(03):77.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 014]
[2]谢 泽,朱建生,李 雯.基于门控循环单元的铁路客票业务流量数据预测[J].计算机技术与发展,2021,31(10):209.[doi:10. 3969 / j. issn. 1673-629X. 2021. 10. 035]
　XIE Ze,ZHU Jian-sheng,LI Wen.Network Traffic Data Forecast of Railway Passenger Ticket Service System Based on Gated Recurrent Unit[J].,2021,31(03):209.[doi:10. 3969 / j. issn. 1673-629X. 2021. 10. 035]
[3]吴长旺,黄 刚,胡婷婷.融合注意力机制的 GNN 推荐算法[J].计算机技术与发展,2022,32(10):7.[doi:10. 3969 / j. issn. 1673-629X. 2022. 10. 002]
　WU Chang-wang,HUANG Gang,HU Ting-ting.GNN Recommendation Algorithm Fused with Attention Mechanism[J].,2022,32(03):7.[doi:10. 3969 / j. issn. 1673-629X. 2022. 10. 002]
[4]殷 齐,丁 飞,朱 跃,等.基于 CNN 与多尺度特征融合的城市交通流预测模型[J].计算机技术与发展,2022,32(10):175.[doi:10. 3969 / j. issn. 1673-629X. 2022. 10. 029]
　YIN Qi,DING Fei,ZHU Yue,et al.An Urban Traffic Flow Prediction Model Based on CNN and Multi-scale Feature Fusion[J].,2022,32(03):175.[doi:10. 3969 / j. issn. 1673-629X. 2022. 10. 029]
[5]蔡宇航,廖光忠.基于改进降噪自编码模型的网络入侵检测[J].计算机技术与发展,2023,33(02):119.[doi:10. 3969 / j. issn. 1673-629X. 2023. 02. 018]
　CAI Yu-hang,LIAO Guang-zhong.Network Intrusion Detection Based on Improved Denoising Autoencoder Model[J].,2023,33(03):119.[doi:10. 3969 / j. issn. 1673-629X. 2023. 02. 018]
[6]赵嘉雨,段亚茹,何立明.基于 GRU-GCN-RDrop 模型的交通速度预测[J].计算机技术与发展,2023,33(04):120.[doi:10. 3969 / j. issn. 1673-629X. 2023. 04. 018]
　ZHAO Jia-yu,DUAN Ya-ru,HE Li-ming.Traffic Speed Prediction Based on GRU-GCN-RDrop Model[J].,2023,33(03):120.[doi:10. 3969 / j. issn. 1673-629X. 2023. 04. 018]
[7]魏 思,李欣泽*,郤丽媛,等.上下文特征注入融合的空气污染物浓度预测[J].计算机技术与发展,2023,33(09):196.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 029]
　WEI Si,LI Xin-ze *,XI Li-yuan,et al.Air Pollutant Concentration Prediction Based on Context Feature Injection and Fusion[J].,2023,33(03):196.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 029]
[8]肖 红,钱祎鸣.基于改进 DenseNet 的固井质量评价新方法[J].计算机技术与发展,2024,34(01):193.[doi:10. 3969 / j. issn. 1673-629X. 2024. 01. 028]
　XIAO Hong,QIAN Yi-ming.A New Method of Cementing Quality Evaluation Based on Improved DenseNet[J].,2024,34(03):193.[doi:10. 3969 / j. issn. 1673-629X. 2024. 01. 028]
[9]欧阳有恒*,严大卫.基于 GRU_LSTM 及 RL 算法的伪随机指令生成器[J].计算机技术与发展,2024,34(02):78.[doi:10. 3969 / j. issn. 1673-629X. 2024. 02. 012]
　OUYANG You-heng*,YAN Da-wei.Pseudo Random Instruction Generator Based on GRU_LSTM and Reinforcement Learning Algorithms[J].,2024,34(03):78.[doi:10. 3969 / j. issn. 1673-629X. 2024. 02. 012]

更新日期/Last Update: 2020-03-10