Building an Open-Source Community to Enhance Autonomic Nervous System Signal Analysis: DBDP-Autonomic

TL;DR

This work addresses the fragmentation and reproducibility barriers in autonomic nervous system signal analysis by proposing an open-source, community-driven framework (DBDP Autonomic) for preprocessing, contextualization, and multimodal data fusion of peripheral signals. It builds on the existing DBDP platform by adding ANS-specific processing, contextual feature extraction, and a GUI-driven, plugin-based architecture to enable transparent, end-to-end pipelines with versioned data-supply chain metadata. It grounds the design in a survey (SUN) highlighting substantial user needs for interoperability and open tools, and outlines integration with Open Science and the Digital Health Data Repository. The proposed approach aims to standardize analyses, improve interpretability, and accelerate the development of digital health biomarkers and interventions through collaborative, open science practices.

Abstract

Smartphones and wearable sensors offer an unprecedented ability to collect peripheral psychophysiological signals across diverse timescales, settings, populations, and modalities. However, open-source software development has yet to keep pace with rapid advancements in hardware technology and availability, creating an analytical barrier that limits the scientific usefulness of acquired data. We propose a community-driven, open-source peripheral psychophysiological signal pre-processing and analysis software framework that could advance biobehavioral health by enabling more robust, transparent, and reproducible inferences involving autonomic nervous system data.

Figures (3)

• Figure 1: This figure demonstrates the need for additional context when analyzing ambulatory physiological signals. We used a stress-prediction model on heart rate variability data to predict a probability of physiological stress/arousal for a person over 24 hours. From a naïve interpretation, it seems there are three major stressful (or high arousal) periods (Fig. 1b). However, when asked to self-report stress levels, the user rated a mix of low to high stress for those periods, contradictory to the purely physiological interpretation (Fig. 1c). Considering additional contextual information, we realize that only one high-arousal episode was stressful since the user was undergoing an exam. The other periods were when the user was in a class and exercising later during the day, which showed similar physiological arousal but were not stressful (Fig. 1d).
• Figure 2: DBDP Autonomic extends the functionalities of DBDP for biobehavioral research. A. As signals from various modalities enter the analysis pipeline, DBDP Autonomic provides additional features on top of the existing modules in DBDP. These features extract and add contextual information, provide domain knowledge (for parameter tuning), and support multimodal data fusion. B. Researchers can then integrate the processed features such as HR, RR, and BP to understand the autonomic constructs in the context of major domains of basic human neurobehavioral functioning. ECG: electrocardiogram, PPG: photoplethysmography, RIP: respiratory inductance plethysmography, BP: blood pressure, EDA: exploratory data analysis, HR: heart rate, HRV: heart rate variability, SBP: systolic BP, DBP: diastolic BP, SCR: skin conductance response, SCL: skin conductance level.
• Figure 3: Community participation model of DBDP-Autonomic. DBDP-Autonomic offers diverse avenues through which users and collaborators can engage, from transmissive actions like 'convey/consume' to transformational activities like 'co-create.' Each stage caters to participants based on their unique goals, skills, and needs, allowing for multifaceted and inclusive community involvement, thus ensuring a richer and more holistic collaboration and contribution. This figure was adapted, with permission, from the Center for Scientific Collaboration and Community Engagement woodley:CSCCECommunityParticipation-2020.