Secure DDI monitoring using QRNG-HCA and BFE-MA-DNN with blockchain integration: a prototype simulation

: The internet of things (IoT) plays a vital role in drug-drug interaction (DDI) monitoring. Unlike traditional approaches lacking security, the proposed framework ensures secure DDI storage using a quantum random number generated key-based Hill cipher algorithm (QRNG-HCA) and a backward feature eliminated maxout activated deep neural network (BFE-MA-DNN) for reliable and secure classification. The proposed framework aims to achieve secure communication for healthcare data, reliable side-channel attack detection, computationally efficient encryption and data processing, and accurate drug-drug interaction prediction to support dependable IoT-enabled healthcare monitoring. Patients equipped with sensor units are modeled as network nodes and grouped into clusters using squared chamfer distance-based fuzzy C-means (SCD-FCM). Cluster heads (CHs) are selected through the TriGen distributed mud ring algorithm (TD-MRA). Each CH employs a BFE-MA-DNN classifier to identify trusted and compromised nodes based on radiation and node features. Data from trusted nodes are collected, pre-processed, and transmitted to the decision controller with encrypted information. The controller classifies DDI outcomes using ResNet-50, then secures the results with QRNG-HCA encryption before storing them on the blockchain. The proposed framework is designed for deployment in hospital IoT environments and remote patient monitoring applications, where continuous drug interaction surveillance, secure medical data transmission, and timely clinical decision support are essential. Its lightweight architecture enables secure and reliable healthcare monitoring across distributed IoT-enabled medical infrastructures. To access and monitor the results, users register with a unique ID and keystroke for authentication, access monitoring results, adjust medication dosages, while a Python-based implementation outperforms existing models experimentally.