• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications >Classification of ERP Signals from Mild Cognitive Impairment Patients with Diabetes using Dual Input Encoder Convolutional Neural Network
【24h】

Classification of ERP Signals from Mild Cognitive Impairment Patients with Diabetes using Dual Input Encoder Convolutional Neural Network

机译：使用双输入编码器卷积神经网络对糖尿病轻度认知障碍患者的ERP信号进行分类

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Mild Cognitive Impairment (MCI) is the early stage of Alzheimers Disease (AD), which is an irreversible neurological disease. Type II diabetes is also closely related to brain recognition, so it is extremely necessary to diagnose the patient's cognition as early as possible. Among them, the event-related potential (ERP) based on electroencephalogram (EEG) is an important research method. In order to diagnose the disease more accurately, machine learning has been widely used. However, deep learning still has a broad application space. Convolutional neural networks (CNN) is high-performance model for deep learning. In this study, we constructed a Daul input encoder convolutional neural network (DIE-CNN) for event related potential (ERP) data with type 2 diabetes. In experiments with multiple models, DIE-CNN performed optimally. The results show that DIE-CNN is an effective method to diagnose the condition of the subject, and the method also retains the expansion space to build a more complex comprehensive evaluation system.
机译：轻度认知障碍（MCI）是阿尔茨海默氏病（AD）的早期阶段，这是一种不可逆的神经疾病。 II型糖尿病也与大脑识别密切相关，因此尽早诊断患者的认知是非常必要的。其中，基于脑电图（EEG）的事件相关电位（ERP）是一种重要的研究方法。为了更准确地诊断疾病，机器学习已被广​​泛使用。但是，深度学习仍然具有广泛的应用空间。卷积神经网络（CNN）是深度学习的高性能模型。在这项研究中，我们构建了一种DAUL输入编码器卷积神经网络（DIE-CNN），用于事件相关电位（ERP）数据，具有2型糖尿病。在具有多种型号的实验中，DIE-CNN最佳地进行。结果表明，DIE-CNN是诊断受试者条件的有效方法，该方法还保留了扩展空间以构建更复杂的综合评估系统。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号