Sensor-Based Detection of Fear and Stress in Lower Back Pain Simulation

Heuler A, Oeler C, Richer R, Rohleder N, Barke A, Weise C (2025)

Publication Type: Conference contribution, Abstract of a poster

Publication year: 2025

Event location: Universität Wien, Österreich AT

Abstract

Background. Chronic primary low back pain is affecting millions globally, significantly impacting daily life. A key factor in this condition is fear avoidance (FA) behavior, in which individuals avoid certain movements for fear of worsening pain. This creates a cycle of physical inactivity, leading to deconditioning, disability, and distress. 

Purpose. This pilot study assesses participants' FA and stress responses during back-stressing movement tasks. We aim to discriminate between high and low FA based on movement patterns during standard tasks using state-of-the-art machine learning (ML) algorithms. The main focus of this study is to investigate feasibility and acceptability of the methodology. 

Method. Twenty pain-free participants performed movement tasks from the Back Performance Scale and a sit-to-stand-task with and without a low back pain simulator. Pain, fear, avoidance, and stress were measured through self-report, psychophysiological, and biological markers (e.g., heart rate variability, salivary cortisol). Movement was assessed with motion capture. In a qualitative interview, participants evaluated feasibility and acceptability.

Results. Analysis of preliminary data revealed higher pain reports during the movement tasks with the back pain simulator than without. Computation of the ML algorithms is still ongoing. Qualitative data indicated principal feasibility and acceptability. Participants noted that the back pain simulator and motion capture suit together restricted their movement, possibly influencing movement outcomes. 

Conclusion. This pilot study supports the feasibility of using a back pain simulator with motion capture to assess pain-related movement behavior. Insights gained will guide technical refinements for the main study focusing on movement, pain, stress, and fear-avoidance classification.

Authors with CRIS profile

Annika Heuler Lehrstuhl für Klinische Psychologie und Behavioral Health Technology Carolin Oeler Lehrstuhl für Klinische Psychologie und Behavioral Health Technology Robert Richer Machine Learning and Data Analytics Lab Nicolas Rohleder Lehrstuhl für Gesundheitspsychologie Cornelia Weise Lehrstuhl für Klinische Psychologie und Behavioral Health Technology

Additional Organisation(s)

Sonderforschungsbereich 1483/1 - Empathokinästhetische Sensorik - Sensortechniken und Datenanalyseverfahren zur empathokinästhetischen Modellbildung und Zustandsbestimmung (SFB EmpkinS)

Related research project(s)

InterFEARence – Assessing feared movements in people with chronic back pain with EmpkinS sensor technology (SFB 1483 - D07) Empatho-Kinaesthetic Sensor Technology (SFB 1483 EmpkinS) Oct. 1, 2024 - July 31, 2027

Involved external institutions

Universität Duisburg-Essen (UDE) DE Germany (DE)

How to cite

APA:

Heuler, A., Oeler, C., Richer, R., Rohleder, N., Barke, A., & Weise, C. (2025, August). Sensor-Based Detection of Fear and Stress in Lower Back Pain Simulation. Poster presentation at International Congress of Behavioral Medicine (ICBM), Universität Wien, Österreich, AT.

MLA:

Heuler, Annika, et al. "Sensor-Based Detection of Fear and Stress in Lower Back Pain Simulation." Presented at International Congress of Behavioral Medicine (ICBM), Universität Wien, Österreich 2025.

BibTeX: Download