Automated Real-time Classification of Functional States Based on Physiological Parameters
An automated real-time classification of human functional states is an important problem for stress resistance evaluation, supervision over operators of critical infrastructure, automated teaching and phobia therapy. In this paper we propose a novel method for binary classification of functional states based on the integrated analysis of (peripheral) physiological parameters: galvanic skin response, respiratory rate, electrocardiographic data, body temperature, electromyographic data, photoplethysmographic data, muscle contraction. The method is based on Gradient Boosted Trees algorithm. A testing of the method showed that in case of stress vs. calm wakefulness differentiation a reliability of the method exceeds 80%.