首页> 外文会议>IEEE International New Circuits and Systems Conference >A Blood Glucose Level Prediction System Using Machine Learning Based on Recurrent Neural Network for Hypoglycemia Prevention
【24h】

A Blood Glucose Level Prediction System Using Machine Learning Based on Recurrent Neural Network for Hypoglycemia Prevention

机译:基于递归神经网络的机器学习血糖水平预测系统预防低血糖症

获取原文

摘要

A blood glucose level prediction system that uses machine learning with a recurrent neural network is presented, to prevents hypoglycemia. By introducing machine learning associated with a recurrent neural network, a high accuracy can be obtained with only a time-series of the blood glucose level. For verifying the proposed method, the system is evaluated from the following viewpoints: the alert frequency, the accuracy rate of the alerts, and the prediction of hypoglycemia after 30 min. The system is able to achieve a prediction accuracy of approximately 80%. In addition, the possibility of a further performance improvement by increasing the frequency of blood glucose measurement is verified. Finally, a method for optimizing the hyper-parameters according to the operating period is presented.
机译:提出了一种使用机器学习和递归神经网络的血糖水平预测系统,以防止低血糖症。通过引入与递归神经网络相关的机器学习,仅通过血糖水平的时间序列就可以获得高精度。为了验证所提出的方法,从以下角度对系统进行了评估:警报频率,警报的准确率以及30分钟后的低血糖预测。该系统能够实现大约80%的预测精度。另外,证实了通过增加血糖测量频率来进一步改善性能的可能性。最后,提出了一种根据运行周期优化超参数的方法。

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