Wan Yuchai, Zheng Zhongshu, Liu Ran, Zhu Zheng, Zhou Hongen, Zhang Xun, Boumaraf Said
Beijing Key Laboratory of Big Data Technology for Food Safety, Beijing Technology and Business University, Beijing 100048, China.
Beijing Lab of Intelligent Information Technology, School of Computer Science, Beijing Institute of Technology, Beijing 100081, China.
Life (Basel). 2021 Jun 18;11(6):582. doi: 10.3390/life11060582.
Many computer-aided diagnosis methods, especially ones with deep learning strategies, of liver cancers based on medical images have been proposed. However, most of such methods analyze the images under only one scale, and the deep learning models are always unexplainable. In this paper, we propose a deep learning-based multi-scale and multi-level fusing approach of CNNs for liver lesion diagnosis on magnetic resonance images, termed as MMF-CNN. We introduce a multi-scale representation strategy to encode both the local and semi-local complementary information of the images. To take advantage of the complementary information of multi-scale representations, we propose a multi-level fusion method to combine the information of both the feature level and the decision level hierarchically and generate a robust diagnostic classifier based on deep learning. We further explore the explanation of the diagnosis decision of the deep neural network through visualizing the areas of interest of the network. A new scoring method is designed to evaluate whether the attention maps can highlight the relevant radiological features. The explanation and visualization make the decision-making process of the deep neural network transparent for the clinicians. We apply our proposed approach to various state-of-the-art deep learning architectures. The experimental results demonstrate the effectiveness of our approach.
已经提出了许多基于医学图像的肝癌计算机辅助诊断方法，尤其是那些采用深度学习策略的方法。然而，大多数此类方法仅在单一尺度下分析图像，并且深度学习模型总是难以解释。在本文中，我们提出了一种基于深度学习的多尺度和多层次融合的卷积神经网络（CNN）方法，用于磁共振图像上的肝脏病变诊断，称为MMF-CNN。我们引入了一种多尺度表示策略，对图像的局部和半局部互补信息进行编码。为了利用多尺度表示的互补信息，我们提出了一种多层次融合方法，将特征级和决策级的信息分层组合，并基于深度学习生成一个强大的诊断分类器。我们通过可视化网络的感兴趣区域，进一步探索深度神经网络诊断决策的解释。设计了一种新的评分方法来评估注意力图是否能突出相关的放射学特征。这种解释和可视化使深度神经网络的决策过程对临床医生来说是透明的。我们将所提出的方法应用于各种最先进的深度学习架构。实验结果证明了我们方法的有效性。
