Feasibility Analysis of AI-Driven Aura Imaging for Early Detection of Stress and Physiological Imbalance

Abstract

Non-invasive health monitoring has become an important area of research, as it enables early identification of physiological and psychological imbalances without the need for invasive procedures. In recent years, there has been growing interest in exploring subtle bioelectromagnetic and thermal emissions of the human body, often interpreted as aura-like patterns, for health assessment. However, the lack of standardized datasets and reliable computational frameworks has limited the scientific validation of such approaches.  This study presents a feasibility analysis of an Artificial Intelligence (AI)-driven aura imaging framework for the early detection of stress and physiological imbalance. A small-scale experimental dataset is developed using controlled image acquisition techniques that incorporate aura-like visual patterns derived from thermal or processed imaging. The collected data is preprocessed to enhance relevant features, followed by feature extraction using Convolutional Neural Networks (CNNs). The extracted features are then used to train classification models for identifying different stress levels. The experimental results demonstrate that even with a limited dataset, AI models can capture meaningful patterns associated with stress-related conditions and achieve satisfactory classification performance. The findings suggest that AI-based aura imaging has the potential to serve as a supplementary, non-invasive tool for early health monitoring. At the same time, the study highlights key challenges, including limited data availability and the need for standardized acquisition protocols, which must be addressed for practical deployment.