Babylonian Journal of Internet of Things

RNN-Based Framework for IoT Healthcare Security for Improving Anomaly Detection and System Integrity

S. Rajaprakash

Department of Computer science & Engineering,Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology , Chennai , Tamilnadu, India

https://orcid.org/0000-0003-2237-5850

C. Bagath Basha

Department of Computer Science and Engineering, Kommuri Pratap Reddy Institute of Technology, Autonomous, Hyderabad, Telangana, India.

https://orcid.org/0000-0002-4622-753X

M. Nithya

Department of Computer Science and Engineering , Vinayaka Missions Kirupananda Variyar Engineering college, Vinayaka Missions Research Foundation, Salem, Tamil Nadu.

https://orcid.org/0009-0008-1055-0686

K. Karthik

Department of Computer Science and Engineering,Aarupadai Veedu Institute of Technology,Chennai 603104, Tamil Nadu, India.

https://orcid.org/0000-0001-5614-479X

Nitisha Aggarwal

Panipat Institute of Engineering and Technology, Samalkha, Haryana, India.

https://orcid.org/0009-0007-3668-5645

S. Kayathri

Department of Computer Science and Engineering, P.S.R Engineering college, Sivakasi, Tamil Nadu 626140, India.

https://orcid.org/0000-0003-0086-2359

DOI: https://doi.org/10.58496/BJIoT/2024/013

Keywords: IoT Healthcare, Recurrent Neural Networks (RNNs), Deep Learning, Classification, Healthcare Data Protection

---

Abstract

The rapid rise of the Internet of Things greatly benefited the healthcare sector by opening up new ways of monitoring patients and ingeniously collecting data on disease management. However, this increased connectivity and health system data interchange introduce many critical security vulnerabilities that are more likely to discredit highly sensitive patient information along with system integrity. The paper investigates only one critical IoT health care security issue, for which a new security framework based on RNNs was used to investigate enhancements in the threat detection and response. This approach modelled network traffic and device behavior sequentially for anomaly and potential breach detection using RNNs. Hence, we introduce the RNN-based model combined with an inclusive security architecture, including data encryption, mechanisms of authentication, and monitoring tools in real time. Experimental results prove that our RNN-based framework significantly improves malicious activity detection and reduces false positives compared to traditional security solutions. The proposed model would provide a strong, scalable, and adaptable security solution tailored to the IoT healthcare environment dynamics. These findings could indicate how RNNs can enhance security in IoTs and provide new ways in which better and more secure healthcare systems can be developed.