Deep Radiomic Analysis Based on Modeling Information Flow in Convolutional Neural Networks
Deep Radiomic Analysis Based on Modeling Information Flow in Convolutional Neural Networks
复制标题
DOI：
10.1109/access.2019.2930238
复制
发表时间：
2019-01-01
期刊：
影响因子：
3.9
通讯作者：
Niazi, Tamim
中科院分区：
文献类型：
--
作者：
Chaddad, Ahmad;Toews, Matthew;Niazi, Tamim
本文提出了一种新的图像特征集的基础上的卷积神经网络（CNN）的原则信息论分析。卷积滤波器的输出被建模为以对象类和网络滤波器组为条件的随机变量。滤波器输出的条件熵（CENT）在理论和实验中被证明是一种高度紧凑和类信息的特征，可以从CNN特征图中计算出来，并用于获得比原始CNN本身更高的分类精度。实验涉及使用3D大脑MRI数据的三个二进制分类任务：阿尔茨海默病（AD）与健康对照（HC）、年轻与老年以及男性与女性，其中CENT特征分类的曲线下面积（AUC）值（93.9%，96.7%和71.9%）显著高于为任务训练的原始CNN分类器的softmax输出（81.6%，79.4%和63.1%）。基于Wilcoxon检验的统计分析确定了与大脑标签有显著联系的CENT特征，这些特征可能作为诊断生物标志物。
This paper proposes a novel image feature set based on a principled information theoretic analysis of the convolutional neural network (CNN). The output of convolutional filters is modeled as a random variable conditioned on the object class and network filter bank. The conditional entropy (CENT) of filter outputs is shown in theory and experiments to be a highly compact and class-informative feature that can be computed from the CNN feature maps and used to obtain higher classification accuracy than the original CNN itself. Experiments involve three binary classification tasks using the 3D brain MRI data: Alzheimer's disease (AD) versus healthy controls (HC), young versus old age, and male versus female, where the area under the curve (AUC) values for the CENT feature classification (93.9%, 96.7%, and 71.9%) are significantly higher than the softmax output of the original CNN classifier trained for the task (81.6%, 79.4%, and 63.1%). A statistical analysis based on the Wilcoxon test identifies CENT features with significant links to brain labels, which could potentially serve as diagnostic biomarkers.