AUTHOR=Raphan Theodore , Yakushin Sergei B. 

TITLE=Predicting Vasovagal Responses: A Model-Based and Machine Learning Approach

JOURNAL=Frontiers in Neurology

VOLUME=12

YEAR=2021

URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2021.631409

DOI=10.3389/fneur.2021.631409

ISSN=1664-2295

ABSTRACT=<p>Vasovagal syncope (<bold>VVS</bold>) or neurogenically induced fainting has resulted in falls, fractures, and death. Methods to deal with <bold>VVS</bold> are to use implanted pacemakers or beta blockers. These are often ineffective because the underlying changes in the cardiovascular system that lead to the syncope are incompletely understood and diagnosis of frequent occurrences of <bold>VVS</bold> is still based on history and a tilt test, in which subjects are passively tilted from a supine position to 20° from the spatial vertical (to a 70° position) on the tilt table and maintained in that orientation for 10–15 min. Recently, is has been shown that vasovagal responses (<bold>VVRs</bold>), which are characterized by transient drops in blood pressure (<bold>BP</bold>), heart rate (<bold>HR</bold>), and increased amplitude of low frequency oscillations in <bold>BP</bold> can be induced by sinusoidal galvanic vestibular stimulation (sGVS) and were similar to the low frequency oscillations that presaged <bold>VVS</bold> in humans. This transient drop in <bold>BP</bold> and <bold>HR</bold> of 25 mmHg and 25 beats per minute (bpm), respectively, were considered to be a <bold>VVR</bold>. Similar thresholds have been used to identify <bold>VVR's</bold> in human studies as well. However, this arbitrary threshold of identifying a <bold>VVR</bold> does not give a clear understanding of the identifying features of a VVR nor what triggers a <bold>VVR</bold>. In this study, we utilized our model of <bold>VVR</bold> generation together with a machine learning approach to learn a separating hyperplane between normal and <bold>VVR</bold> patterns. This methodology is proposed as a technique for more broadly identifying the features that trigger a <bold>VVR</bold>. If a similar feature identification could be associated with <bold>VVRs</bold> in humans, it potentially could be utilized to identify onset of a <bold>VVS</bold>, i.e, fainting, in real time.</p>