Editorial: Body-centric computing for health and wellbeing

Zhuying Li, Xipei Ren, Chengyu Liu, Ding Ding, Xinyi Fu

Abstract

This editorial synthesizes five research contributions exploring the integration of digital technology with human physiology. By examining diverse applications—from e-textile pain communication for patients with cognitive impairments to AI-driven exercise prediction—the collection highlights a paradigm shift toward "body-centric computing." This field seeks to create proactive, personalized interventions that support physical and mental health through seamless, multi-modal technological integration.

Introduction

While wearable technology is ubiquitous, much of it remains "extrinsic"—devices we strap on that provide raw data but don't necessarily integrate into our lived experience or provide actionable, personalized interventions. The authors argue that the current challenge is moving beyond mere health monitoring toward body-centric computing.

This field addresses the gap between human physiology and digital technology. It seeks to solve problems like:

  • Providing feedback without overwhelming the user’s cognitive load.
  • Using virtual environments (VR) to increase physical exertion.
  • Accurately detecting emotions using multiple physiological signals.
  • Providing a "voice" for those who cannot communicate their physical pain, such as elderly patients with cognitive impairments.

Methodological Synthesis

As an editorial, this paper synthesizes the methodologies of five distinct research contributions:

  • User Experience (UX) Research: Investigating the "sense of agency" and cognitive load in wearable design through varied feedback patterns.
  • VR Exergaming Experiments: Testing the impact of Non-Player Character (NPC) spectators on player heart rate and engagement in gesture-based VR games.
  • Multi-modal Signal Processing: Developing Emotion Recognition Systems (ERS) that fuse data from photoplethysmography (PPG), galvanic skin response (GSR), and skin temperature (SKT).
  • Machine Learning (ANNs): Utilizing Artificial Neural Networks to predict long-term exercise persistence based on a seven-week walking program.
  • Hardware Prototyping (E-textiles): Embedding physiological sensors into everyday objects (handkerchiefs) to detect pain-indicative physical activities.

Data Sources

While the editorial does not name specific public repositories, it describes the data sources used across the five studies:

  • Physiological Signals: Real-time streams of heart rate, skin conductance, and body temperature.
  • Longitudinal Activity Data: A dataset tracking participants over a seven-week physical activity program.
  • Clinical Markers: Integration of BMI and cholesterol levels into predictive models.
  • VR Performance Metrics: Data on player exertion, heart rate percentage, and subjective game experience scores.

Key Insights and Discussion

  • The Feedback Paradox: Research by Morris et al. found that feedback deviating from expected patterns increases user awareness but can simultaneously reduce the "sense of agency" and increase cognitive load. This suggests that "more information" isn't always better; design must balance alertness with mental bandwidth.
  • Social Presence in VR: Xu et al. demonstrated that the presence of spectator NPCs in VR exergames isn't just a visual flourish—it significantly increases physical exertion and heart rate, suggesting that "social" pressure (even from AI) is a powerful motivator for health.
  • The Power of Multi-modality: Gohumpu et al. showed that emotion recognition is significantly more accurate when combining multiple peripheral signals (like skin temp and heart rate) rather than relying on a single metric, which is often prone to noise.
  • Clinical Data as a Predictor: Diaconau et al. found that adding traditional clinical markers like BMI and cholesterol to neural network models significantly improved the accuracy of predicting whether a person would stick to an exercise routine.
  • Dignity through E-Textiles: Li et al. introduced a novel application for e-textiles: using sensor-embedded handkerchiefs to help Alzheimer’s patients communicate chronic pain non-verbally, bridging a critical communication gap in elder care.

Results

  • Emotion Recognition: The multi-modal ERS developed by Gohumpu et al. outperformed existing single-modal systems in accuracy.
  • Exercise Persistence: The inclusion of BMI and cholesterol data into ANN models provided a more robust method for anticipating physical activity persistence.
  • VR Engagement: Studies showed that a larger number of spectator NPCs in VR games led to higher average heart rate percentages and improved overall game experience.

Limitations and Future Work

The authors acknowledge that while these technologies are promising, several hurdles remain:

  • Seamless Integration: There is a persistent need for technologies that integrate into daily life without being intrusive or requiring significant user effort.
  • Personalization: Future systems must better adjust to individual requirements and environmental settings to be truly effective.
  • Mental Health Focus: The authors call for more research into how interactive technology can specifically bolster mental wellbeing and stress relief, such as through augmented reality (AR) breathing exercises.
  • Algorithm Development: There is a continuing need for more sophisticated algorithms that can translate raw biosensor data into meaningful health interventions.

Summary updated

Abstract

Body-centric computing is rapidly emerging as a transformative approach for enhancing health and 2 wellbeing. This field focuses on the integration of computational technologies with the human body, aiming 3 to monitor, analyze, and improve various aspects of physical and mental health (Mueller et al., 2018).At the core of body-centric computing are technologies such as wearable sensors, smart fabrics, and 5 integrated devices that gather crucial health data in real time (Li et al., 2023). These technologies often 6 track one's health metrics and encourage active wellness management (Ren et al., 2018). Its applications 7 extend beyond mere health monitoring to include devices that support healthy living, enhance medical 8 rehabilitation, personalize treatments, and improve overall wellbeing through immersive experiences. For 9 example, wearables can measure exercise intensity, heart rate changes, and mood fluctuations to offer 10 personalized health advice, encouraging users to adjust their lifestyles for improved well-being. Another 11 example includes augmented reality glasses that facilitate stress-relieving breathing exercises within natural 12 environment settings, aiming to lower anxiety and bolster mental health (Zheng et al., 2023).In recent years, interest in body-centric computing for health and wellbeing has increased, with a growing 14 emphasis on developing user-friendly technologies that seamlessly integrate into daily life. The ultimate aim 15 is to create a symbiotic relationship between the human body and digital technology, enabling individuals 16 to proactively manage their health and wellbeing with informed, data-driven decisions, thereby offering a 17 new paradigm in health and wellness management.In this research topic, we call for submissions that explore the use of body-centric computing for health In total, five papers were eventually accepted in this Research Topic. These studies offer a diverse array of 26 insights ranging from design implications and health monitoring to practical applications, each pushing the 27 boundaries of how we integrate technology with the human body for enhanced living. Below, we provide a 28 brief overview of these innovative contributions.
BibTeX
@article{li2024editorial,
  author = {Zhuying Li and Xipei Ren and Chengyu Liu and Ding Ding and Xinyi Fu},
  title = {{Editorial: Body-centric computing for health and wellbeing}},
  journal = {Frontiers of Computer Science},
  year = {2024},
  doi = {10.3389/fcomp.2024.1393102},
}
Figure