Thyroid Nodule Classification in Ultrasound Images by Fine-Tuning Deep Convolutional Neural Network.

Thyroid Nodule Classification in Ultrasound Images by Fine-Tuning Deep Convolutional Neural Network.
复制标题

超声图像中的甲状腺结节分类通过微调深卷积神经网络。

DOI:
10.1007/s10278-017-9997-y
复制
发表时间:
2017-08
影响因子:
4.4
通讯作者:
Eramian M
Eramian M
中科院分区:
工程技术2区
文献类型:
--
作者:
Chi J;Walia E;Babyn P;Wang J;Groot G;Eramian M

文献摘要

参考文献

被引文献

相似文献

由于许多甲状腺结节是偶然发现的,因此通过细针穿刺活检或手术排除那些极有可能是良性结节的同时,尽可能多地识别恶性结节是很重要的。本文介绍了一种用于甲状腺结节超声图像分类的计算机辅助诊断系统。我们使用深度学习方法从甲状腺超声图像中提取特征。对超声图像进行预处理以校准其尺度并去除伪影。然后使用预处理的图像样本对预训练的GoogLeNet模型进行微调,从而获得更好的特征提取。将提取的甲状腺超声图像特征发送给成本敏感随机森林分类器,将图像分为“恶性”和“良性”两类。实验结果表明,改进后的GoogLeNet模型对开放存取数据库(Pedraza et al.)图像的分类精度为98.29%,灵敏度为99.10%,特异度为93.90%,对本地卫生区域数据库的分类精度为96.34%,灵敏度为86%,特异度为99%。
With many thyroid nodules being incidentally detected, it is important to identify as many malignant nodules as possible while excluding those that are highly likely to be benign from fine needle aspiration (FNA) biopsies or surgeries. This paper presents a computer-aided diagnosis (CAD) system for classifying thyroid nodules in ultrasound images. We use deep learning approach to extract features from thyroid ultrasound images. Ultrasound images are pre-processed to calibrate their scale and remove the artifacts. A pre-trained GoogLeNet model is then fine-tuned using the pre-processed image samples which leads to superior feature extraction. The extracted features of the thyroid ultrasound images are sent to a Cost-sensitive Random Forest classifier to classify the images into “malignant” and “benign” cases. The experimental results show the proposed fine-tuned GoogLeNet model achieves excellent classification performance, attaining 98.29% classification accuracy, 99.10% sensitivity and 93.90% specificity for the images in an open access database (Pedraza et al.), while 96.34% classification accuracy, 86% sensitivity and 99% specificity for the images in our local health region database.
DOI: 10.4015/s1016237210001815
发表时间: 2010-04-01
影响因子: 0.9
作者:
Chang, Chuan-Yu;Liu, Hsiang-Yi;Shih, Shyang-Rong
通讯作者: Shih, Shyang-Rong
DOI: 10.1148/radiol.11110206
发表时间: 2011-09-01
期刊: RADIOLOGY
影响因子: 19.7
作者:
Kwak, Jin Young;Han, Kyung Hwa;Kim, Eun-Kyung
通讯作者: Kim, Eun-Kyung
DOI: 10.1007/978-3-319-10404-1_65
发表时间: 2014
期刊: LECTURE NOTES IN ARTIFICIAL INTELLIGENCE
影响因子: --
作者:
Roth, Holger R.;Lu, Le;Seff, Ari;Cherry, Kevin M.;Hoffman, Joanne;Wang, Shijun;Liu, Jiamin;Turkbey, Evrim;Summers, Ronald M.
通讯作者: Summers, Ronald M.
DOI: 10.1177/0954411912472422
发表时间: 2013-01-01
影响因子: 1.8
作者:
Acharya, U. Rajendra;Sree, S. Vinitha;Suri, Jasjit S.
通讯作者: Suri, Jasjit S.
DOI: 10.1016/j.ultrasmedbio.2010.08.019
发表时间: 2010-12-01
影响因子: 2.9
作者:
Chen, Shao-Jer;Chang, Chuan-Yu;Wei, Chang-Kuo
通讯作者: Wei, Chang-Kuo