System and method for allergic reaction detection

Abstract

The present disclosure provides a system and a method for predicting identifying an allergic reaction of a subject, using sensors and learning. The benefit of this system is in its ability to recognize an allergic reaction even before the visible symptoms appear. It relies on cardiovascular-related parameters, which are not difficult to generate, and can be produced by simple sensors, such as a photoplethysmography (PPG) sensor and an ECG sensor and features which are extracted from these parameters, such as heart rate variability (HRV). One of the challenges that the system of the present invention offers a solution for is handling the false positive cases, i.e., physiological reactions which may correspond with allergic reaction in terms of the measured parameters but are not in fact resulted from actual allergic reactions. This may occur, for example, when the subject is engaged in physical activity. To deal with these cases, the present invention incorporates sophisticated algorithms, which provide the ability to adjust the deviation of the parameter, and specifically HRV, from the baseline, in two time-frames scales, short-term (a time-window in the scale of minutes, e.g. 5 minutes) and long-term (a time-window that is not sensitive to, and therefore masks, momentarily or short-term peaks, such as physical exercise. The time-window is typically in the scale of about 24 hours but can be less or more than 24 hours).

Dependencies are inferred from the claim text.