Wearable Sensor-Based Intervention of Postural Risk during Fluoroscopic Procedures
Loading...
Authors
Bonin, William R
Date
2025-08-27
Type
thesis
Language
eng
Keyword
Work Related Musculoskeletal Disorders , Posture , Ergonomics , RULA , Exposure Variation Analysis , Interventional Cardiology , Orthopaedic Trauma , Vibrotactile , Wearable Sensors , Physician , Biomechanics
Alternative Title
Abstract
Work-related musculoskeletal disorders (WMSDs) are highly prevalent among healthcare workers, particularly those involved in interventional and surgical procedures. Sustained awkward posture, prolonged task durations and the usage of heavy lead aprons for radiation protection contributes to chronic strain, notably in the neck and lower back. Traditional ergonomic assessment tools such as the Rapid Upper Limb Assessment Tool (RULA) are limited in their ability to capture to continuous, task-specific nature of clinical work. This thesis presents a two-part investigation aimed at addressing this limitation through wearable sensing and real- time feedback. The first study introduces a cohort investigation in Interventional Cardiology and Orthopaedic Trauma procedures collected from 47 procedures to capture spinal posture. Using real-time RULA scoring, this work identified that nurses and physicians spend more than 50% of procedural time in medium-to high-risk postures (RULA scores ≥ 5). Secondary video analysis identified seven recurring task categories as key contributors to maladaptive postures, defined as POSTURE (Pressure, Operations, Technology, Uneven Demographics, Reaching, Exceptions) as an interpretable model for risk analysis. Building on these findings, a second study developed a wearable vibrotactile feedback system that compares traditional RULA-based feedback algorithm with a novel Exposure Variation Analysis (EVA) approach. EVA incorporates posture severity and exposure duration to provide feedback when a cumulative loading score is reached, rather than discrete posture angles. In a simulated surgical task, the EVA algorithm reduced medium-to high-risk postures by 91.5%, compared to 68.5% with RULA-based feedback when normalized by task duration. EVA-based feedback also required fewer cues on average (4.63 ± 2.13 vs. 10.37 ± 13.92) with no significance on perceived cognitive workload and no increase to task duration. Together, these studies demonstrate an ergonomic system that combines continuous posture assessment with algorithm-driven cumulative loading feedback to address WMSD risk in healthcare settings. Future work will focus system miniaturization, improvement of feedback algorithms, integration into training environments and on clinical validation. By promoting healthier working habits and supporting long-term healthcare worker well-being, this research has the potential to reduce WMSD-related injuries, extend career longevity, and improve overall quality of life of healthcare professionals.