基于脑机接口的嗅觉神经信号识别筛查阿尔茨海默症的研究  

The Recognition and Screening for Alzheimer's Disease Based on Olfactory Neural Signal from Brain Computer Interface

作　　者：段言 王天星 袁群琛 石颖倩 胡燕婕[4] 庄柳静[1,2] 王平[1,2] DUAN Yan;WANG Tianxing;YUAN Qunchen;SHI Yingqian;HU Yanjie;ZHUANG Liujing;WANG Ping(Biosensor National Special Laboratory,College of Biomedical Engineering and Instrument Science,Zhejinag University,Hangzhou Zhejiang 310027,China;Key Laboratory for Biomedical Engineering of Education Ministry,College of Biomedical Engineering and Instrument Science,Zhejiang University,Hangzhou Zhejiang 310027,China;Zhejiang e-Linkcare Meditech Co.,Ltd.,Hangzhou Zhejiang 310011,China;Run Run Shaw Hospital,Zhejiang University School of Medicine,Hangzhou Zhejiang 310016,China)

机构地区：[1]浙江大学生仪学院生物传感器国家专业实验室,浙江杭州310027 [2]浙江大学生仪学院生物医学工程教育部重点实验室,浙江杭州310027 [3]浙江亿联康医疗科技有限公司,浙江杭州310011 [4]浙江大学医学院附属邵逸夫医院,浙江杭州310016

出　　处：《传感技术学报》2024年第9期1650-1658,共9页Chinese Journal of Sensors and Actuators

基　　金：国家自然科学基金项目(62271443,32250008);浙江省自然科学基金项目(LBY21H180001,LY21C100001)。

摘　　要：本文研究了一种基于脑机接口的嗅觉神经信号筛查阿尔茨海默症的方法。针对哺乳动物嗅觉系统的生理结构,设计开发了植入式柔性神经电极,将电极植入小鼠嗅觉系统,读取在体嗅觉神经信号。以阿尔茨海默症模型小鼠为研究对象,提取模型小鼠和对照小鼠的自发嗅觉神经活动,分析了神经信号在时域和频域上的分布差异,结果表明阿尔茨海默症小鼠嗅觉系统中存在异常的神经振荡。此外,提出了一种基于嗅觉神经信号的阿尔茨海默症分类识别与筛查方法。对嗅觉神经信号进行特征提取,利用多种机器学习模型实现了阿尔茨海默症的分类,其中ANN模型准确率最高,达到89.98%。实验结果表明,所研究的方法为AD疾病的筛查与诊断提供了一种新的技术手段。A novel method for Alzheimers disease(AD)recognition utilizing olfactory neural signals based on brain-computer interface in proposed.Flexible neural electrodes are developed and implanted into the olfactory system of mice to record in vivo neural activities.Time and frequency domain analysis of the spontaneous signals from the AD model mice and control group reveals that there are abnor-mal neural oscillations in the AD mice.Moreover,classification models are proposed to recognize AD.Several machine learning classifi-ers are constructed and trained using features extracted from the olfactory neural signals.The ANN model has the highest accuracy rate of 89.98%.The results provide a new technique method with great potential for AD screening and diagnosis.

关 键 词：生物医学工程 阿尔茨海默症诊断 嗅觉神经信号 机器学习 

分 类 号：R741.044[医药卫生—神经病学与精神病学] TP181[医药卫生—临床医学] R749.16[自动化与计算机技术—控制理论与控制工程]