Artificial Intelligence Techniques for Secure and Energy-Efficient Secure MRI Image Transmission via IoT Devices and Hybrid Physics-Guided Neural Networks: Trends and Challenges

Abstract

The integration of Artificial Intelligence with Internet of Things (IoT) technologies has significantly transformed modern healthcare systems, particularly in the secure transmission of Magnetic Resonance Imaging (MRI) data. MRI images contain highly sensitive patient information and require robust security mechanisms along with efficient transmission techniques due to their large size. This paper presents a comprehensive review of AI-based techniques for secure and energy-efficient MRI image transmission using IoT devices and hybrid physics-guided neural networks. Recent advancements highlight the use of deep learning models such as Convolutional Neural Networks (CNNs), Generative Adversarial Networks (GANs), and hybrid architectures combined with encryption techniques to ensure data confidentiality and integrity. Studies demonstrate that integrating deep learning with encryption algorithms such as chaotic and Arnold transformations allows secure MRI processing without compromising diagnostic accuracy. Additionally, IoT-based healthcare systems utilizing wireless sensor networks and MQTT protocols enable real-time and low-power data transmission, making them suitable for remote monitoring applications. Furthermore, physics-guided neural networks have emerged as a promising approach by incorporating domain-specific knowledge into deep learning models, improving reconstruction accuracy and reducing data dependency. These models enhance robustness and efficiency, particularly in resource-constrained IoT environments. However, challenges such as computational complexity, scalability, and real-time implementation persist. This review consolidates recent developments, identifying key trends, challenges, and future research directions. The integration of AI-driven secure transmission frameworks with IoT systems and physics-guided learning models is expected to play a critical role in developing next-generation intelligent healthcare systems.