Assessment of the electromagnetic waves effects on the SAR of human body based on the random forest machine learning algorithm

With the rapid proliferation of communication technologies and the pervasive presence of electromagnetic fields (EMF) in the living environment, concerns regarding environmental impacts and human health have been raised. Furthermore, the expanding adoption of wireless technologies and the Internet of Things (IoT) has significantly increased human exposure to elevated levels of electromagnetic radiation. The precise assessment of the environmental effects of these waves and their impact on living tissues constitutes a primary challenge in environmental health and biomedical engineering. This research investigates the biophysical mechanisms through which electromagnetic waves interact with human tissues and reviews current safety standards. The primary objective is to propose practical strategies for minimizing exposure and protecting the human body against non-ionizing radiation. Findings indicate that adherence to established safety protocols, implementation of passive protective measures, increasing physical distance from radiation sources, and the use of shielding equipment can effectively reduce the risk of both thermal and non-thermal biological damage. Conventional approaches for measurement and numerical simulation—such as the Finite-Difference Time-Domain (FDTD) method—are often time-consuming and resource-intensive. After the training phase, AI-based methods respond with exceptionally high speed and incur very low computational costs; however, their accuracy depends on the quality and volume of training data, and they may exhibit errors under unknown conditions. This paper explores the application of Artificial Intelligence (AI) and Machine Learning (ML) as powerful tools for predicting the Specific Absorption Rate (SAR) and evaluating risks associated with electromagnetic waves.