Abstract：Longitudinal data are generated in the epidemiological follow-up study, but the analysis of such data is always a difficult point in statistics. At present, we mostly use the traditional linear mixed model to analyze such longitudinal data. The model is strict in the distribution of data, moreover, it is assumed that the disease is a linear progression, its application would be subject to some restrictions. Some researchers have proposed using the neural network model to analyze longitudinal follow-up data in recent years. This review discusses the current research status of neural network model in longitudinal follow-up data analysis and provides a new way for the analysis of longitudinal follow-up data.
张文茜， 苏海霞， 孙丽君， 张玉海. 神经网络模型在纵向随访数据分析中的研究进展[J]. 实用预防医学, 2017, 24(1): 127-129.
ZHANG Wen-qian, SU Hai-xia, SUN Li-jun, ZHANG Yu-hai. Research progress on neural network model in longitudinal follow-up data analysis. , 2017, 24(1): 127-129.
[1] 庄严.纵向数据的实验设计及统计分析理论[J].数理医药学杂志, 2011,24(1):75-77. [2] Hedeker D, Gibbons RD. Longitudinal data analysis[M]. New Jersey: John Wiley & Sons,2006:105-108. [3] 李会民,王普,方丽英,等.基于混合效应模型的纵向数据建模方法研究[J].内蒙古大学学报(自然科学版), 2014,45(1):79-83. [4] Laird NM, Ware JH. Random-effects models for longitudinal data[J]. Biometrics 1982:963-974. [5] Fishbaugh J, Durrleman S, Piven J, et al. A framework for longitudinal data analysis via shape regression[J].Proc Spie, 2012,(3):208-220. [6] 张磊,孔桂兰,马谢民. 人工神经网络在糖尿病住院费用研究中的应用[J].中国医院管理,2015,35(1):61-64. [7] Bishop CM. Pattern recognition and machine learning[M]. New York:Springer,2006:66-78. [8] Baxt WG. Use of an artificial neural network for data analysis in clinical decision-making: the diagnosis of acute coronary occlusion[J]. Neural Comput, 1990, 2(4): 480-489. [9] Da Silva Lopes HF, Abe JM, Anghinah R. Application of paraconsistent artificial neural networks as a method of aid in the diagnosis of Alzheimer disease[J]. J Med Syst, 2010, 34(6): 1073-1081. [10] Pappada SM, Cameron BD, Rosman PM, et al. Neural network-based real-time prediction of glucose in patients with insulin-dependent diabetes[J]. Diabetes Technol The, 2011, 13(2):135-141. [11] 韩敏,席剑辉,范明明. 神经网络应用于多元变量时间序列的建模研究[J]. 仪器仪表学报, 2006,19(3): 275-279. [12] 陈权.BP神经网络在结直肠癌预后研究中的应用[D].武汉:华中科技大学, 2011. [13] 杨同满,郭雨. 基于遗传算法的BP神经网络时间序列预测算法及其应用[J]. 电脑知识与技术, 2015,31(1): 160-162. [14] 丁静静,王阿明,巩萍. BP神经网络在脑胶质瘤患者术后3年生存期预测中的应用[J]. 徐州医学院学报, 2015, 35(6):415-417. [15] Longhi S, Nijkamp P, Maierhofer E. Neural network modeling as a tool for forecasting regional employment patterns[J]. Intern Reg SciRev, 2005, 28 (3):330-346. [16] Patuelli R, Reggiani A, Nijkamp P, et al. New neural network methods for forecasting regional employment: an analysis of German labour markets[J]. Spatial Econ Anal, 2006,1(1):7-30. [17] Maity TK, Pal AK. Subject specific treatment to neural networks for repeated measures analysis[J]. Proceed Internl MultiConfEng Comp Scient, 2013, 1(1):60-65. [18] Tandon R, Adak S, Kaye JA. Neural networks for longitudinal studies in Alzheimer's disease[J]. Artif Intell Med, 2006,36(3):245-255.
2017, Vol. 24 
