CT2US: Cross-modal transfer learning for kidney segmentation in ultrasound images with synthesized data

计算机科学 人工智能 学习迁移 分割 试验装置 深度学习 卷积神经网络 数据集 超声波 模式识别(心理学) 试验数据 人工神经网络 放射科 医学 程序设计语言
作者
Yuxin Song,Jing Zheng,Long Lei,Zhipeng Ni,Baoliang Zhao,Ying Hu
出处
期刊:Ultrasonics [Elsevier BV]
卷期号:122: 106706-106706 被引量:84
标识
DOI:10.1016/j.ultras.2022.106706
摘要

Accurate segmentation of kidney in ultrasound images is a vital procedure in clinical diagnosis and interventional operation. In recent years, deep learning technology has demonstrated promising prospects in medical image analysis. However, due to the inherent problems of ultrasound images, data with annotations are scarce and arduous to acquire, hampering the application of data-hungry deep learning methods. In this paper, we propose cross-modal transfer learning from computerized tomography (CT) to ultrasound (US) by leveraging annotated data in the CT modality. In particular, we adopt cycle generative adversarial network (CycleGAN) to synthesize US images from CT data and construct a transition dataset to mitigate the immense domain discrepancy between US and CT. Mainstream convolutional neural networks such as U-Net, U-Res, PSPNet, and DeepLab v3+ are pretrained on the transition dataset and then transferred to real US images. We first trained CNN models on a data set composed of 50 ultrasound images and validated them on a validation set composed of 30 ultrasound images. In addition, we selected 82 ultrasound images from another hospital to construct a cross-site data set to verify the generalization performance of the models. The experimental results show that with our proposed transfer learning strategy, the segmentation accuracy in dice similarity coefficient (DSC) reaches 0.853 for U-Net, 0.850 for U-Res, 0.826 for PSPNet and 0.827 for DeepLab v3+ on the cross-site test set. Compared with training from scratch, the accuracy improvement was 0.127, 0.097, 0.105 and 0.036 respectively. Our transfer learning strategy effectively improves the accuracy and generalization ability of ultrasound image segmentation model with limited training data.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
刚刚
刚刚
刚刚
橙子完成签到,获得积分10
1秒前
噜噜噜的应助被科研通管家采纳,获得10
1秒前
情怀的应助被科研通管家采纳,获得10
2秒前
2秒前
小蘑菇的应助被科研通管家采纳,获得10
2秒前
2秒前
搜集达人的应助被科研通管家采纳,获得10
2秒前
Owen的应助被科研通管家采纳,获得10
2秒前
情怀的应助被科研通管家采纳,获得10
2秒前
Jun的应助被科研通管家采纳,获得10
2秒前
今后的应助被科研通管家采纳,获得10
3秒前
思源的应助被科研通管家采纳,获得10
3秒前
TSW发布了新的文献求助10
3秒前
3秒前
传奇3的应助被科研通管家采纳,获得10
3秒前
捶捶自己完成签到,获得积分10
3秒前
彭于晏的应助被科研通管家采纳,获得10
3秒前
ccc关注了科研通微信公众号
3秒前
FashionBoy的应助被科研通管家采纳,获得10
3秒前
wanci的应助被清秀豪英采纳,获得10
3秒前
酷波er的应助被科研通管家采纳,获得10
3秒前
Jun的应助被科研通管家采纳,获得10
3秒前
null的应助被科研通管家采纳,获得10
4秒前
大模型的应助被科研通管家采纳,获得10
4秒前
大模型的应助被郭姗姗采纳,获得10
4秒前
4秒前
4秒前
4秒前
hzy完成签到 ,获得积分10
5秒前
diuwaitao发布了新的文献求助10
5秒前
6秒前
6秒前
蛋黄完成签到,获得积分10
7秒前
tlc_191026发布了新的文献求助10
7秒前
7秒前
于之淇完成签到,获得积分20
8秒前
HaoranMa发布了新的文献求助10
8秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
Improving the wear resistance of vehicle gear oil by oil soluble LaF3 nanoparticles and typical additives of gear oil 500
Boundaries and Bodhisattvas 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8183508
求助须知:如何正确求助?哪些是违规求助? 9572862
关于积分的说明 21827639
捐赠科研通 7780681
什么是DOI,文献DOI怎么找? 3434016
关于科研通互助平台的介绍 2558189
邀请新用户注册赠送积分活动 2480405