医学图像深度学习技术:从卷积到图卷积的发展  

Deep learning-based medical images analysis evolved from convolution to graph convolution

作　　者：唐朝生[1,2] 胡超超 孙君顶 司马海峰[1] Tang Chaosheng;Hu Chaochao;Sun Junding;Sima Haifeng(School of Computer Science&Technology,Henan Polytechnic University,Jiaozuo 454000,China;Provincial Key Laboratory for Computer Information Processing Technology,Soochow University,Suzhou 215006,China)

机构地区：[1]河南理工大学计算机科学与技术学院,焦作454000 [2]苏州大学江苏省计算机信息处理技术重点实验室,苏州215006

出　　处：《中国图象图形学报》2021年第9期2078-2093,共16页Journal of Image and Graphics

基　　金：河南省科技厅科技攻关项目(212102310084);苏州大学江苏省计算机信息处理技术重点实验室开放课题(KJS2048)。

摘　　要：以卷积神经网络为代表的深度学习技术推动神经网络在医学图像研究领域不断实现新突破。然而,平移不变性等理论假设限制了卷积神经网络在非欧氏空间数据中的表达能力,是医学图像深度学习技术亟待突破的瓶颈。图卷积技术不仅能够解决非欧氏空间数据的拓扑建模难题,还实现了空间特征提取,是深度学习技术全新的研究方向。本文对图卷积网络在医学图像领域的相关理论及其应用进行综述,旨在系统归纳和全面总结医学图像领域最新的图卷积理论、方法和实践,包括图结构视角下医学图像的专业采集、数据结构的剪枝转换以及特征聚类重构方法;图卷积网络的理论溯源,重要的网络架构和发展脉络;图卷积网络的优化方向和衍生出的跳跃连接、inception、图注意力等重要机制;图卷积网络在医学图像分割、疾病检测和图像重建等方面的实践应用。最后,提出了图卷积网络在医学图像分析领域仍亟待突破的瓶颈问题:1)多模态医学图像学习中,异构图的构建与学习任务的优化;2)特征重构和池化过程中,如何通过构图算法设计与神经架构搜索算法结合,以实现最优图结构的可学习过程转换;3)高质量图结构医学标注数据的大规模低成本生成与生成对抗网络的算法设计。随着人工智能技术的不断发展和医学影像规模的不断扩大,以图卷积为代表的深度学习方法必将在医疗辅助诊断领域取得更大的突破。The convolutional neural networks(CNN) have been facilitated to develop deep learning-based medical image sustainable research.The translation invariance capability has constrained the expression of CNN in the context of nonEuclidean spatial data.In order to realize deep learning-based spatial feature extraction, graph convolution has resolved the topology modeling issue based on non-Euclidean spatial data.The latest theories and applications of graph convolutional networks(GCN) for medical image analysis have been reviewed.This research has been divided into four aspects as follows: 1) Data structure transformation of medical images based on graph-structure;2) Theoretical development and network architecture of GCN;3) The optimized and derivative of graph convolution mechanism;4) GCN implementation in medical image segmentation, disease detection, and image reconstruction.First, graph-structure-based medical images transformation has been reviewed in the context of graph data acquisition, transformation, and reconstruction.The graphstructure-based medical data have been acquired via the professional medical equipment, the sparse pruning algorithm, or the rebuilt graph-structure using the K-nearest neighbor( KNN) algorithm.The graph-structure reconstruction algorithm based on the medical image features has performed better than the graph-structure conversion algorithm based on the medical image data.Next, the critical architecture of the GCN, including the graph convolutional layer, the graph regularization layer, the graph pooling layer, and the graph readout layer, has been summarized.The graph-structural nodes or edges have been updated via the graph convolution layer.The generalization of GCN has been upgraded via the graph regularization layer.The number of calculation parameters has been reduced via the graph pooling layer.The representation of the graph has been generated via the graph readout layer.Graph convolution has been categorized into two methods as mentioned below: a) The spectrum-based graph conv

关 键 词：医学图像 深度学习 图表示学习 图神经网络(GNN) 图卷积网络(GCN) 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]