The study addresses the cyber threats and data breaches to which connected healthcare devices can be exposed. The authors state that existing cybersecurity mechanisms often fail to ensure the integrity, confidentiality, and availability of healthcare data, while noisy datasets and limited computational resources complicate attack detection and prevention. Their proposed architecture combines machine-learning-based classification of normal and attack patterns with blockchain-based controls for authenticated, tamper-resistant, and reliable data exchange in cloud environments.

Before classification, the framework applies a preprocessing procedure called Adjusted Min–Max with Decimal Scaling and Statistical Column Normalization, abbreviated AMM-DSSCN. The method is designed to handle inconsistencies and scale features in the two datasets. A separate Sharpbelly Fish Optimization algorithm then performs feature selection by retaining features identified as most relevant.

Threat classification is conducted with Nested Dense Graph Attention Capsule Networks, or NDGACN, while Circulatory System-Based Optimization tunes the model’s weight parameters. For data sharing, the researchers integrated a blockchain-based Generation Right Trading model with digital smart contracts. On CICIoMT2024, NDGACN-CSBO achieved an accuracy of 99.23%, an F1-score of 99.72%, and a reported ROC value of 99.20%.

On WUSTL EHMS 2020, the corresponding results were 99.88% accuracy, 99.56% F1-score, and 99.56% ROC value. The study also reports security integrity above 99%. These measurements describe performance in the authors’ experimental evaluations of the two datasets, rather than a described deployment in live hospitals or clinical environments.

The available evidence is limited to the study’s two datasets and reported experimental setup. It also provides no information in the available material about implementation costs, latency, hardware requirements, interoperability with particular clinical platforms, or behavior against previously unseen attacks. The results therefore show strong benchmark performance for the proposed IoHT security framework, but they do not establish that the same outcomes will occur in every healthcare network or clinical setting.

Subject of Research: Experimental evaluation of a blockchain-integrated cybersecurity framework for detecting attacks against Internet of Healthcare Things systems.

Article Title: Blockchain-oriented IoHT cybersecurity system utilizing nested dense graph attention capsule networks with circulatory system-based optimization for robust data protection

Article References: Dinesh, G., Prasad, R. G., Guganathan, L., & Anupama, J. (2026). Blockchain-oriented IoHT cybersecurity system utilizing nested dense graph attention capsule networks with circulatory system-based optimization for robust data protection. Knowledge and Information Systems, 68(1), Article 293. https://doi.org/10.1007/s10115-026-02877-0

Image Credits: AI Generated

DOI: 10.1007/s10115-026-02877-0

Keywords: Internet of Healthcare Things, cybersecurity, blockchain, threat detection, machine learning, healthcare data, smart contracts, capsule networks, graph attention, data protection, Blockchain-oriented, IoHT