Decoding Vagal Activity Across Scales: AI and Machine Learning Approaches to Autonomic Function

The vagus nerve lies at the intersection of cardiovascular, neuroimmune, metabolic, and behavioral regulation, exerting influence across multiple levels of physiological organization. From single-fiber compound action potentials to network-level rhythms such as heart rate variability (HRV), vagal pathways provide measurable signatures of autonomic balance and adaptive state. Yet the translation of these signals into robust, clinically actionable predictors of health and disease remains fragmented. Preclinical electrophysiology, clinical HRV analytics, and closed-loop vagus nerve stimulation (VNS) each generate high-dimensional, richly patterned data — but these data streams often exist in isolation, limiting synthesis of principles that span scales, species, and experimental paradigms. This gap highlights the need for integrative frameworks capable of uniting autonomic physiology across the molecular, neural, and systemic domains.This Research Topic aims to identify and model the shared organizing principles that govern vagal and autonomic dynamics across scales, leveraging recent advances in computational neuroscience, network physiology, and artificial intelligence. For the purposes of this Research Topic, we use “artificial intelligence (AI)” as the broader category encompassing learned, rule-based, and hybrid computational systems applied to autonomic data, and “machine learning (ML)” to refer specifically to data-driven approaches — including classical statistical learning, deep learning, and foundation-model architectures — that infer structure or predictions directly from physiological recordings. By connecting vagal electroneurogram (VENG) recordings, HRV-derived features, and VNS dose–response data with clinically relevant outcomes, this Topic seeks to bridge fundamental neurophysiology with translational application. The goal is to develop mechanistically grounded, generalizable predictors that can inform diagnosis, track recovery, and guide intervention across a spectrum of conditions — including perinatal adaptation, cardiovascular dysfunction, inflammation, sepsis, and neuropsychiatric disorders. Through this synthesis, the Topic aims to unify traditionally disconnected domains — from single-axon recordings to population-level predictive models — within the conceptual framework of autonomic regulation.To advance this integrative agenda in autonomic neuroscience, we welcome contributions addressing, but not limited to, the following themes:AI approaches (such as machine learning, deep learning, and foundation models) for HRV and autonomic signal analysis in clinical outcome predictionDirect vagal recordings (VENG): fiber-selective neurography, classification, and decoding approachesClosed-loop VNS and biomarker-guided stimulation paradigmsVagal regulation of neuroimmune function and computational signatures of inflammatory toneAutonomic regulation of cardiovascular function across development and agingCross-scale and network-physiology approaches linking molecular, cellular, and systems-level dynamicsInformation-theoretic, energy-based, or complexity frameworks for autonomic predictionTranslational pipelines and benchmarking for autonomic biomarkers and model reproducibilityReal-world and wearable monitoring of autonomic signals with AI-driven analyticsInterpretability and generalizability of AI/ML models for autonomic signal analysisMethods-led contributions without a clear autonomic mechanism, and AI/ML papers using autonomic signals purely as substrate without mechanistic insight, are outside the scope of this Research Topic.This Research Topic welcomes Original Research, Methods, Brief Reports, Systematic Reviews, Reviews, Perspectives, Hypothesis & Theory papers, Clinical Trials, and Data Reports.Dr. Martin Gerbert Frasch holds US Patents Nos. WO2025217002, US20110152633, US11622710, relating to fetal and maternal health monitoring. He holds equity in, and serves as advisor of, Delfina Care, Inc, and nurtur health, Inc, both active in pregnancy health. These interests are not directly related to the subject of this work. Topic editors: Martin Gerbert Frasch, Christopher G Wilson Source: https://www.frontiersin.org/research-topics/81514/decoding-vagal-activity-across-scales-ai-and-machine-learning-approaches-to-autonomic-function