Dual Secure Optimal Trusted Routing for Sensitive Data Transfer to Ensure Accurate Patient Healthcare State Prediction Using IoT-Enabled Wireless Sensor Networks

Abstract

With the rapid advancement of Internet of Things (IoT) and Wireless Sensor Networks (WSNs), healthcare systems have evolved to support continuous patient monitoring, real-time data acquisition, and cloud-based decision support. The secure transmission of sensitive medical data and the reliability of healthcare decision-making remain major challenges. Traditional routing techniques fail to provide robust trust management, making the system vulnerable to malicious nodes and unreliable data paths. The lack of lightweight, end-to-end encryption increases the risk of data breaches during transmission. Compounding the issue is the limited diagnostic accuracy of conventional analytics platforms, which struggle to effectively process complex, high-dimensional healthcare data. To address this, this study introduces a Dual Secure optimal Trusted routing (DST-Route) technique designed to ensure secure, trust-aware data transfer and enhance patient diagnostic decision-making in IoT-WSN. In the data transfer phase, the Enhanced Pomarine Jaeger Optimization (EPJO) algorithm is used to perform trust-based clustering and optimal cluster head selection, ensuring that only reliable nodes participate in data transmission. The sensitive health data collected from patients is protected using SmartNetcryption, a lightweight encryption used to secure information before cloud storage. In the analytics phase, the framework uses pre-trained deep learning models, including ResNet, DenseNet, EfficientNet, and UNet for feature extraction, while a Modular Deep Transfer Learning (MDTL) enables accurate healthcare state prediction and early diagnosis. Experimental results demonstrate that DST-Route significantly improves trust accuracy, energy efficiency, and prediction performance when compared to conventional routing techniques. The proposed UNet, combined with the MDTL model, achieved a healthcare state prediction accuracy of 98% with a loss rate of 0.05, showing 12.54% improvement over state-of-the-art models. This performance underscores the effectiveness of the DST-Route technique in ensuring secure and reliable sensitive data transfer for accurate patient state prediction.