Assessing the Level of Autonomic Nervous Activity for Effective Biofeedback Training

Abstract: This paper proposes a prototype of a new biofeedback training based on mathematical models of cardiovascular control. For this purpose we develop a low-cost device that is able to record and process arterial pulse wave via photoplethysmograph and skin temperature on the peripheral part of the arm. A benchmark analysis of our device against a registered cardiovascular measurement system (Biopac MP35 from Biopac Inc., USA) shows that heart rate and skin temperature values delivered by our device are acceptable with a very low residual error (+/-1 beat/min, +/-1 oC). Both measured signals are feed into a mathematical model which estimates the level of activity of both sympathetic and parasympathetic branches of the autonomic nervous system. That information is used jointly with other biological parameters for investigating the stress score of the subject. The data is processed in real-time and continuously displayed to the user for effective biofeedback training. The complete solution was preliminary tested on three volunteers who used the displayed biofeedback information in order to regulate their emotional state successfully during Biofeedback training. They exhibited a significant reduction in stress score compared to three control subjects who did not used our solution during biofeedback training. This supports the benefits of biofeedback training with autonomic nervous tone assessment as effective holistic healing method.