论文部分内容阅读
目的:比较反向传播算法(BP)神经网络和径向基函数(RBF)神经网络预测老年痴呆症疾病进展的效果。方法:以老年痴呆症随访数据为研究对象,以性别、年龄、受教育程度、有无高血压、有无高胆固醇、有无心脏病、有无中风史、有无家族史8个指标作为输入变量,以五年随访的MMSE差值为输出变量,构建基于BP神经网络和RBF神经网络的老年痴呆症疾病进展预测模型。结果:与BP神经网络模型相比,RBF神经网络预测的结果更好,能够有效地预测老年痴呆症疾病进展。结论:神经网络模型将老年痴呆症疾病进展预测问题转化为随访数据中相关测量指标与MMSE差值的非线性问题,为复杂的老年痴呆症疾病进展预测提供了新思路。
Objective: To compare the effects of backpropagation (BP) neural network and radial basis function (RBF) neural network in predicting the progression of Alzheimer’s disease. Methods: The follow-up data of Alzheimer’s disease were used as the research object. Eight variables including gender, age, education level, with or without high blood pressure, with or without high cholesterol, heart disease, history of stroke, and family history were taken as input variables The MMSE difference of five years follow-up was taken as the output variable to build a predictive model of Alzheimer’s disease progression based on BP neural network and RBF neural network. Results: Compared with BP neural network model, RBF neural network can predict better results and effectively predict the progress of Alzheimer’s disease. CONCLUSION: The neural network model translates the prediction of Alzheimer’s disease progression into the non-linearity of the correlation measure and the MMSE difference in the follow-up data, providing new insights into the prediction of complex Alzheimer’s disease progression.