Abstract Big data and artificial intelligence develop rapidly. Big data analysis has been applied in many fields of smart healthcare. Once these data are leaked or modified during transmission, it will not only invade the privacy of patients, but also endanger their lives. Many researchers worked on encrypted personal health records (PHR). However, there are still some challenges, such as data leakage during deep learning and leakage of training models, and some users do not want their data to be leaked to the analysis organization. How to protect privacy while leveraging deep learning is a pressing issue. In this paper, we present a system for predicting disease and timely alarms by collecting data from sensors and using deep learning to analyze and monitor patient health data. In order to protect the privacy of health data, we adopt an assured data deletion approach which the data owner can choose to revoke some users’ access to their health data. Extensive analysis and experimental results are presented that demonstrate the security, and efficiency of our proposed approach.

中文翻译：

---

一种用于可穿戴无线传感器网络的隐私保护大数据分析系统

摘要 大数据和人工智能发展迅速。大数据分析已经应用于智慧医疗的多个领域。这些数据一旦在传输过程中被泄露或修改，不仅会侵犯患者的隐私，还会危及患者的生命。许多研究人员致力于加密个人健康记录 (PHR)。但是，仍然存在一些挑战，例如深度学习过程中的数据泄露和训练模型的泄露，并且一些用户不希望他们的数据泄露给分析机构。如何在利用深度学习的同时保护隐私是一个紧迫的问题。在本文中，我们提出了一种通过从传感器收集数据并使用深度学习来分析和监控患者健康数据来预测疾病和及时警报的系统。为了保护健康数据的隐私，我们采用有保证的数据删除方式，数据所有者可以选择撤销部分用户对其健康数据的访问权限。提供了广泛的分析和实验结果，证明了我们提出的方法的安全性和效率。