文章摘要

文章信息/Info

作者:: 廖华龙1; 曾小茜2; 李华凤3; 于洋4; 赵灿1; 陈宇1△; 1.四川大学生物力学工程省重点实验室，成都 610065；2.四川大学华西医院华西大数据中心，成都 610041；3.四川大学华西第二医院麻醉科，成都 610041；4.四川大学华西医院肾脏内科，成都 610041

Author(s):: LIAO Hualong1; ZENG Xiaoxi2; LI Huafeng3; YU Yang4; ZHAO Can1; CHEN Yu1; 1.Laboratory of Biomechanics Engineering, Sichuan University, Chengdu610065, China;2. West China Big Data Center, West China Hospital, Sichuan University, Chengdu 610041;3.Department of Anesthesiology, West China Second University Hospital, Sichuan University, Chengdu 610041;4.Department of Nephrology, West China Second University Hospital, Sichuan University, Chengdu 610041

Keywords:: Machine learning; Diseases; Prediction; Algorithm; Feature selection

摘要:: 本文通过阐述机器学习的定义及分类，介绍了K近邻、朴素贝叶斯、逻辑回归、支持向量机、决策树、集成的树模型以及人工神经网络等研究者常用于医疗疾病预测的机器学习算法，重点分析了机器学习在预测心血管疾病、糖尿病、肾病、肿瘤、妊娠期疾病等几类常见疾病中的应用情况，主要从特征选择、算法和预测准确性等方面说明机器学习预测疾病的应用特点。

Abstract:: We expound the concept and classification of machine learning and introduce several common kinds of machine learning methods applied on medical disease prdiction practice, such as K-nearest neighbor, Naive Bayes, logistic regression, support vector machine, decision tree, integrated tree models and artificial neural networks，we focus on the machine learning application in the prediction of some common diseases, such as cardiovascular diseases, diabetes, kidney diseases, tumour, pregnancy diseases and so on. We mainly analyze the characteristics of machine learning in disease prediction from three aspects: feature selection, algorithm and model accuracy.