A conceptual analysis for the measurement of stress intensity by deep learning using EEG signals

Several Health-related issues are caused due to mental stress. In Initial stages the level of mental stress can be measured using tools which are developed by Physicians and Scientists. Many neuroimaging tools have been suggested to measure the stress developed in the workplace. Electroencephalogram (EEG) signals are a key component as they contain vast knowledge related to conditions and states of mind. Existing methods for EEG Signal analysis are discussed to measure level of stress in the paper. This review focuses the crucial variations in-between the research results and debates those fluctuations from the data analysis methods give rise to many contrary results. The fluctuations in results may be because of several factors like no proper protocol, the area of interest within the brain, stressor types, experiment period, EEG analysis, methodology used by which properties are extracted, and classifier type. Hence, the major factor relating to detection stress in mind, by choosing the major suitable properties. A convoluted come varied band of properties relating to EEG, with variation in time, working, and dynamic connections of brain, required inclusion of different methods to learn about their connections with mental stress. Subsequently, the evaluation recommends merging cortical activations with the integration to the network calculates with approaches of deep learning to enhance the correctness range of mental stress calculation.