Use of a convolutional neural network and quantitative ultrasound for diagnosis of fatty liver.
Use of a convolutional neural network and quantitative ultrasound for diagnosis of fatty liver.
复制标题
使用卷积神经网络和定量超声诊断脂肪肝。
DOI：
10.1016/j.ultrasmedbio.2020.10.025
复制
发表时间：
2021-03
影响因子：
2.9
通讯作者：
Oelze ML
中科院分区：
文献类型：
--
作者：
Nguyen TN;Podkowa AS;Park TH;Miller RJ;Do MN;Oelze ML
定量超声（QUS）被用来分类兔子，诱导有肝脏疾病，将它们放在一个定义的时间内的脂肪饮食和/或定期注射四氯化碳。肝脏状态的基础事实是基于通过Folch测定和羟脯氨酸浓度估计的脂肪肝浓度，以量化纤维化。使用SonixOne扫描仪和L9 - 4/38线性阵列对家兔进行体内超声扫描。使用QUS或1D卷积神经网络（CNN）基于来自肝脏的超声背散射射频（RF）信号对肝脏脂肪百分比进行分类。使用QUS参数与线性回归和典型相关分析（CCA）表明，QUS参数可以区分肝脏脂质水平高于或低于5%。然而，QUS参数对纤维化不敏感。CNN通过分析原始RF超声信号来实现，而不使用单独的参考数据。CNN将肝脏的分类输出为高于或低于肝脏中5%脂肪水平的阈值。CNN优于利用QUS参数联合收割机与支持向量机（SVM）在区分低和高脂质肝脏水平方面的分类，即，测试数据的准确率分别为74%和59%。因此，虽然CNN没有提供组织特性的物理解释，例如，由于介质或散射体特性的衰减，CNN在预测脂肪肝状态方面具有高得多的准确性，并且不需要外部参考扫描。
Quantitative ultrasound (QUS) was used to classify rabbits that were induced to have liver disease by placing them on a fatty diet for a defined duration and/or periodically injecting them with CCl4. The ground truth of the liver state was based on lipid liver percents estimated via the Folch assay and hydroxyproline concentration to quantify fibrosis. Rabbits were scanned ultrasonically in vivo using a SonixOne scanner and an L9–4/38 linear array. Liver fat percentage was classified based on the ultrasonic backscattered radio-frequency (RF) signals from the livers using either QUS or a 1D convolutional neural network (CNN). Use of QUS parameters with linear regression and canonical correlation analysis (CCA) demonstrated that the QUS parameters could differentiate between livers with lipid levels above or below 5%. However, the QUS parameters were not sensitive to fibrosis. The CNN was implemented by analyzing raw RF ultrasound signals without using separate reference data. The CNN output the classification of liver as either above or below a threshold of 5% fat level in the liver. The CNN outperformed the classification utilizing the QUS parameters combine with a support vector machine (SVM) in differentiating between low and high lipid liver levels, i.e., accuracies of 74% versus 59% on the testing data. Therefore, while the CNN did not provide a physical interpretation of the tissue properties, e.g., attenuation of the medium or scatterer properties, the CNN had much higher accuracy in predicting fatty liver state and did not require an external reference scan.
登录
查看更多内容
影响因子：
2.4
作者：
Oelze, ML;Zachary, JF
通讯作者：
Zachary, JF
影响因子：
13.5
作者：
Ziol, M;Handra-Luca, A;Beaugrand, M
通讯作者：
Beaugrand, M
影响因子：
0.9
作者：
Choong CC;Venkatesh SK;Siew EP
通讯作者：
Siew EP
DOI：
10.1111/1440-1681.12102
发表时间：
2013-07-01
影响因子：
2.9
作者：
Kubota, Norihiro;Kado, Shoichi;Ishikawa, Fumiyasu
通讯作者：
Ishikawa, Fumiyasu
影响因子：
2.9
作者：
Nguyen, Trong N.;Podkowa, Anthony S.;Oelze, Michael L.
通讯作者：
Oelze, Michael L.