IoHT-SHIELD: Blockchain-Assisted Anomaly Detection Framework for Securing IoHT Devices

Abstract

With the rapid proliferation of Internet of Healthcare Things (IoHT), cybersecurity challenges have emerged in the form of resource-constrained and interconnected devices. To overcome these challenges, this project proposes a blockchain-based anomaly detection system that ensures intelligent threat detection, tamper-proof logging, and secure data management. The system utilizes machine learning algorithms such as Isolation Forest, SVM, CatBoost, and LSTM Autoencoders to identify anomalies in data streams generated by Internet of Healthcare Things devices. The system securely logs the anomalies in the Ethereum blockchain and stores data in IPFS to ensure data integrity and decentralization. The system utilizes FastAPI to establish communication with the system and process data in real-time. The results of the proposed system show excellent accuracy and improved security performance against benchmark datasets. The proposed system ensures data security in Internet of Healthcare Things environments in a reliable and transparent manner. Index Terms—Internet of Healthcare Things (IoHT), anomaly detection, blockchain, Ethereum, IPFS, CatBoost, Isolation Forest, LSTM Autoencoder, cybersecurity, data integrity.

Introduction

The text explains a cybersecurity framework for Internet of Health Things (IoHT) systems that combines machine learning and blockchain technology to improve security, reliability, and data integrity in healthcare environments.

It highlights that IoHT devices generate large amounts of sensitive medical and sensor data, but are vulnerable to cyberattacks due to their interconnected nature and limited security mechanisms. Traditional methods like centralized intrusion detection and conventional encryption are often insufficient because they suffer from scalability issues, high computational cost, and lack of real-time performance.

To address these problems, the proposed system introduces a Blockchain-Assisted Anomaly Detection Framework. It uses multiple machine learning models (such as Isolation Forest, SVM, CatBoost, and LSTM Autoencoders) to detect abnormal behavior in IoHT data in real time. Detected anomalies trigger alerts and are securely recorded on a blockchain, ensuring tamper-proof and transparent logging.

The architecture is organized into several layers: data acquisition, preprocessing, machine learning-based anomaly detection, decision and alert generation, blockchain logging, decentralized storage, and a visualization dashboard. This layered design ensures efficient data handling, real-time threat detection, and secure storage.

Conclusion

For the current paper, we have chosen IoHT-SHIELD technology as the topic of our study. \"IoHT-SHIELD\" stands for a methodology that utilizes the use of the blockchain approach and focuses on the detection of potential risks associated with the safety of IoT-H devices. It goes without saying that security concerns connected with the protection of IoT-H devices should never be ignored. Two points are crucial when it comes to securing IoT-H devices. Firstly, it is crucial to identify any threat. Secondly, it is vital to log all events associated with such threats safely. For the purpose of identifying threats to the IoT-H device, the approach of IoHT-SHIELD technology utilizes various machine learning techniques such as Isolation Forest and SVM, CatBoost and LSTM Autoencoders. Blockchain approach in IoT-H devices uses the technologies of Ethereum blockchain and IPFS respectively.People tested IoHT-SHIELD. It worked well. It found threats accurately. It kept events safe. This is important for healthcare because healthcare has a lot of rules. IoHT-SHIELD is also flexible. It can work with many different Internet of Healthcare Things devices. It can process data in time and it keeps a clear record of what happens. In the future, our efforts will focus on improving IoHT-SHIELD by adding features that clarify its functionality. These updates will include testing with device data, incorporating various sensor data types, and optimising the blockchain component for faster performance.The primary aim of IoHT-SHIELD is to protect IoHT devices, which it accomplishes by detecting devices and maintaining secure records. This enables early threat detection and keeps device data secure. The system is valuable for monitoring IoHT devices, analysing data, and ensuring their proper operation. Overall, IoHT-SHIELD is essential for maintaining the security of IoHT devices.

References

[1] M. Islam, A. S. Dukyil, S. Alyahya, and S. Habib, “An IoT Enabled Anomaly Detection System for Smart City Surveillance using 2D-CNN and Echo State Networks,” Sensors, vol. 23, no. 4, 2023. [Online]. Available: https://www.mdpi.com/1424-8220/23/4/2358 [2] K. DeMedeiros, A. Hendawi, and M. Alvarez, “A Survey of AI-Based Anomaly Detection in IoT and Sensor Networks,” Sensors, vol. 23, no. 3, 2023. [Online]. Available: https://www.mdpi.com/1424-8220/23/3/1352 [3] A. Chatterjee and B. Ahmed, “IoT Anomaly Detection Methods and Applications: A Survey,” arXiv preprint, 2022. [Online]. Available: https://arxiv.org/abs/2207.09092 [4] M. I. H. Okfie et al., “Anomaly Detection in IIoT Transactions using Machine Learning and Blockchain,” Engineering, Technology & Applied Science Research, 2024. [Online]. Available: https://www.etasr.com/index.php/ETASR/article/view/7384 [5] “A Survey of Anomaly Detection in IoT Networks Using Machine Learning,” International Journal of Creative Research Thoughts (IJCRT), 2024. [Online]. Available: https://www.ijcrt.org/papers/IJCRT2402509.pdf [6] M. Domb, “Anomaly Detection in IoT: Recent Advances, AI and ML Techniques,” IntechOpen, 2023. [Online]. Available: https://www.intechopen.com/chapters/87783 [7] M. M. Khan et al., “Anomaly Detection in IoT-Based Healthcare Using Machine Learning,” PubMed, 2024. [Online]. Available: https://pubmed.ncbi.nlm.nih.gov/38467709/ [8] J. Jot and L. S. Sharma, “Anomaly Detection in IoT Sensors using Machine Learning,” IJRASET, 2023. [Online]. Available: https://doi.org/10.22214/ijraset.2023.55226 [9] H. Zahan et al., “IoT-AD: A Framework to Detect Anomalies Among Interconnected IoT Devices,” arXiv preprint, 2023. [Online]. Available: https://arxiv.org/abs/2306.06764 [10] T. Lai et al., “Ensemble Learning based Anomaly Detection for IoT Cybersecurity,” arXiv preprint, 2023. [Online]. Available: https://arxiv.org/abs/2307.10596

Copyright

Copyright © 2026 Muneebullah Hussaini Syed, Mohammad Saif, Mr. B. J. Job Karuna Sagar. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.