基于BP神经网络与行为学损伤的阿尔茨海默病预测模型构建与分析
Journal: Frontier Forum of Clinical Medicine DOI: 10.32629/ffcr.v4i2.19972
Abstract
目的：探讨将行为学损伤特征与BP神经网络相结合用于阿尔茨海默病(AD)患者预测的可行性与性能分析,并构建与验证相应的预测模型。方法：本研究基于阿尔茨海默病神经影像计划(ADNI)数据库,选取242名受试者作为研究对象,其中阿尔茨海默病(AD)106例、认知正常(CN)136例。所有受试者均具备完整行为学评分及人口学信息。以BP神经网络为主要分类器,采用交叉验证进行模型训练、超参数调优与稳健性评估；此外,本研究基于SHAP方法对BP神经网络模型的特征重要性进行了可视化分析。结果：模型的分类性能较好,准确率为84.5%。SHAP分析显示,冷漠(NPIG)对AD预测贡献最高,是模型中最重要的特征。结论：基于BP神经网络的行为学损伤特征预测模型可较稳定、准确地识别阿尔茨海默病风险,其中冷漠、抑郁、食欲与进食障碍等行为学指标具有重要预测价值。
Keywords
BP神经网络；阿尔茨海默病；行为学损伤；预测模型
Full Text
PDF - Viewed/Downloaded: 0 TimesReferences
[1] Kawahara J,Brown C J,Miller S P,et al.BrainNetCNN: Conv
[2] Monastero R,Mangialasche F,Camarda C,et al.A systema
[3] 宋媛媛,陈兵,刘静静,等.BP神经网络模型与Logistic回归模型在精神分裂症患者复发影响因素分析中的比较[J].护理学报,2022,29(13):59⁃65.
[4] Nair V,Hinton G E.Rectified linear units improve restr
[5] Srivastava N, Hinton G, Krizhevsky A,et al.Dropout: a simple way to prevent neural networks from overfitting[J]. The journal of machine learning research,2014,15(1):1929-
[6] Robert P, Lanctôt K L, Agüera-Ortiz L, et al. Is it time to revise the diagnostic criteria for apathy in brain disorders? The 2018 international consensus group[J].European Psychiatry,
[2] Monastero R,Mangialasche F,Camarda C,et al.A systema
[3] 宋媛媛,陈兵,刘静静,等.BP神经网络模型与Logistic回归模型在精神分裂症患者复发影响因素分析中的比较[J].护理学报,2022,29(13):59⁃65.
[4] Nair V,Hinton G E.Rectified linear units improve restr
[5] Srivastava N, Hinton G, Krizhevsky A,et al.Dropout: a simple way to prevent neural networks from overfitting[J]. The journal of machine learning research,2014,15(1):1929-
[6] Robert P, Lanctôt K L, Agüera-Ortiz L, et al. Is it time to revise the diagnostic criteria for apathy in brain disorders? The 2018 international consensus group[J].European Psychiatry,
Copyright © 2026 刘淑娟, 陈迟到, 黄忠浩
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License