随机森林在医学影像分析中的应用研究进展

Research progress on application of random forest in medical image analysis

随机森林(random forest,RF)是一种机器学习方法,在医学、生物信息、管理学等领域广泛应用,常用于处理分类和回归问题。随机森林属于集成学习算法族,特点是在训练过程中加入了数据样本扰动和输入属性扰动,因此可以处理多种数据类型。在现有医学影像分析中,随机森林主要用于以下3个方面:医学图像的图像处理、辅助医学治疗诊断、探究某些病症的发病因素。本文首先对随机森林的基本原理进行简单介绍,然后对随机森林在医学影像中的使用加以重点介绍,最后对随机森林的优缺点加以小结和展望。

Random forest（RF） is a machine learning method widely used in medical science,bioinformatics and management science to deal with classification and regression problems.RF belongs to the family of integrated learning algorithms,which are characterized as adding data sample perturbations and input property perturbations during training to handle a variety of data types.In the existing medical image analysis,RF is mainly used in image processing of medical images,diagnosis of assisted medical treatment,and exploration of the pathogenesis of certain diseases.This paper firstly introduces the basic principle of RF briefly and then focuses on the use of RF in medical imaging.Finally,the advantages and disadvantages of RF are summarized and prospected.