[1]邸敬,马帅,王国栋,等.基于改进Unet与动态阈值可变FCMSPCNN的医学图像分割[J].中国医学物理学杂志,2023,40(3):328-335.[doi:DOI:10.3969/j.issn.1005-202X.2023.03.011]
 DI Jing,MA Shuai,WANG Guodong,et al.Medical image segmentation using improved Unet combined with dynamic threshold changed FCMSPCNN[J].Chinese Journal of Medical Physics,2023,40(3):328-335.[doi:DOI:10.3969/j.issn.1005-202X.2023.03.011]
点击复制

基于改进Unet与动态阈值可变FCMSPCNN的医学图像分割()
分享到:

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
40卷
期数:
2023年第3期
页码:
328-335
栏目:
医学影像物理
出版日期:
2023-03-29

文章信息/Info

Title:
Medical image segmentation using improved Unet combined with dynamic threshold changed FCMSPCNN
文章编号:
1005-202X(2023)03-0328-08
作者:
邸敬马帅王国栋廉敬
兰州交通大学电子与信息工程学院, 甘肃 兰州 730070
Author(s):
DI Jing MA Shuai WANG Guodong LIAN Jing
School of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China
关键词:
CoA Unet脉冲耦合神经网络注意力机制肝脏分割视网膜血管分割
Keywords:
Keywords: CoA Unet pulse coupled neural network attention mechanism liver segmentation retinal vascular segmentation
分类号:
R318;TP391.9
DOI:
DOI:10.3969/j.issn.1005-202X.2023.03.011
文献标志码:
A
摘要:
针对深度学习的医学图像分割模型训练时间长和精度不精的问题,提出结合动态阈值可变FCMSPCNN的多尺度上下文编解码结构和注意力机制的CoA Unet(Context Attention Unet)分割方法。首先,使用动态阈值可变的FCMSPCNN预分割出目标矩形区域并使用掩码遮盖背景部分;然后,更深层卷积块加入快捷连接交叉融合不同层次的特征,并通过注意力门突出对目标特征的学习;最后,在编解码器最底层加入改进的多尺度上下文提取器可以更好地提取目标特征信息。模型分别在LiTs和DRIVE数据集上进行验证,肝脏分割指标Miou、Aver_HD、Aver_Dice分别为0.890 5、6.369 9、0.947 7,视网膜血管分割指标分别为0.589 2、9.255 9、0.740 9。实验表明,预处理能缩短4.3%~20.33%的训练时间并提升2%~6%分割精度,与其他5种分割方法相比,CoA Unet能取得更好的整体分割性能。
Abstract:
Abstract: Aiming at the long training time and low precision of medical image segmentation model based on deep learning, a segmentation method combining the multi-scale context encoding and decoding structure of dynamic threshold changed FCMSPCNN with context attention Unet (CoA Unet) is proposed. Dynamic threshold changed FCMSPCNN is used to pre-segment the target rectangle and mask the background. The deeper convolution block with shortcut connection integrates features at different levels, and highlights the learning of target features through the attention gate. Furthermore, an improved multi-scale context extractor is added to the bottom layer of the codec to better extract target feature information. The model is verified on LiTs and DRIVE datasets, respectively. The Miou, Aver_HD and Aver_Dice of the proposed method are 0.890 5, 6.369 9, 0.947 7 for liver segmentation, and 0.589 2, 9.255 9, 0.740 9 for retinal vascular segmentation. The experiment reveals that the preprocessing can shorten the training time by 4.30%-20.33% and improve the segmentation accuracy by 2%-6%. Compared with other 5 segmentation methods, CoA Unet can achieve better overall segmentation performance.

相似文献/References:

[1]田娟秀,刘国才. 基于NSST变换和PCNN的医学图像融合方法[J].中国医学物理学杂志,2018,35(8):914.[doi:DOI:10.3969/j.issn.1005-202X.2018.08.010]
 TIAN Juanxiu,LIU Guocai. Medical image fusion method based on non-subsampled shearlet transform and pulse coupled neural network[J].Chinese Journal of Medical Physics,2018,35(3):914.[doi:DOI:10.3969/j.issn.1005-202X.2018.08.010]

备注/Memo

备注/Memo:
【收稿日期】2022-11-19 【基金项目】国家自然科学基金(62061023, 61941109);甘肃省科技计划资助项目(22JR5RA360);甘肃杰出青年基金(21JR7RA345) 【作者简介】邸敬,硕士,副教授,主要研究方向:数字图像处理、移动通信关键技术、无线宽带技术,E-mail: 46891771@qq.com
更新日期/Last Update: 2023-03-29