摘要
医疗大数据指数目庞大、增长迅速、结构复杂、隐藏价值高的数据。机器学习技术能够有效分析医疗大数据的内部联系,对疾病的早期诊断及预后具有重要临床指导意义。阐述了机器学习技术在医疗大数据中的应用及研究进展,包括在大数据分析中的回归分析、决策树、基于内核的算法、降低维度算法等浅层机器学习算法模型,卷积神经网络、循环神经网络、自动编码器、深度信念网络等深度学习算法模型,以及各个算法模型的临床应用,分析了机器学习在医疗数据挖掘中的应用前景和存在的技术难题。
Medical big data is a kind of data with huge amount,rapid growth,complex structure and high hidden value.Machine learn ing(ML)technology can effectively analyze and interpret the internal relations of medical big data,and has important clinical guiding significance in the early diagnosis and prognosis of diseases.This article expounds the machine learning technology in the medical ap plication of big data and its research progress,including the analysis of large data in regression analysis,decision tree,based on the kernel algorithm,to reduce the dimension of the shallow machine learning algorithms,circulating neural network model and the convo lution neural network,the automatic encoder,deep deep learning algorithms such as belief network model and the clinical application of each algorithm model,the medical application prospect of machine learning and the existing technical problems in data mining are analyzed.
作者
孙涛
徐秀林
SUN Tao;XU Xiu-lin(College of Medical Devices and Food,Shanghai University of Science and Technology,Shanghai 200093,China)
出处
《软件导刊》
2019年第11期10-14,共5页
Software Guide
关键词
医疗大数据
机器学习
诊断及预后
深度学习
临床应用
medical big data
machine learning
diagnosis and prognosis
deep learning
clinical application
