2021

Automation in Construction

Automated detection and classification of construction workers’ loss of balance events using wearable insole pressure sensors

Maxwell Fordjour Antwi-Afari, Heng Lia, JoonOh Seo, Arnold Yu Lok Wong

Dept. of Building and Real Estate, Hong Kong Polytechnic University

Keywords

Insole pressure sensors Loss of balance, Supervised machine learning

Abstract

Fall on the same level is the leading cause of non-fatal injuries in construction workers; however, identifying loss of balance events associated with specific unsafe surface conditions in a timely manner remain challenging. The objective of the current study was to develop a novel method to detect and classify loss of balance events that could lead to falls on the same level by using foot plantar pressure distributions data captured from wearable insole pressure sensors. Ten healthy volunteers participated in experimental trials, simulating four major loss of balance events (e.g., slip, trip, unexpected step-down, and twisted ankle) to collect foot plantar pressure distributions data. Supervised machine learning algorithms were used to learn the unique foot plantar pressure patterns, and then to automatically detect loss of balance events. We compared classification performance by varying window sizes, feature groups and types of classifiers, and the best classification accuracy (97.1%) was achieved when using the Random Forest classifier with all feature groups and a window size of 0.32 s. This study is important to researchers and site managers because it uses foot plantar pressure distribution data to objectively distinguish various potential loss of balance events associated with specific unsafe surface conditions. The proposed approach can allow practitioners to proactively conduct automated fall risk monitoring to minimize the risk of falls on the same level on sites.

Moticon's Summary

In this study the authors aimed to develop a method to detect and classify loss of balance events which may leas to subsequent falls. For example, falls are the primary cause of non-fatal injuries in construction workers making which exemplifies the relevance of the identification of associated events to enhance occupational safety. For the development of the method plantar pressure data was collected from ten health volunteers who participated in trials simulating four different loss of balance events. In this context plantar pressure data was collected using Moticon sensor insoles. Plantar pressure data served as an input for a supervised machine learning algorithm who was set to detect and classify loss of balance events. Subsequently, classifications performance was evaluated using varying time windows, feature groups and types of classifiers. Results showed that the best classification performance was achieved using a random forest classifier including all feature groups and a time window of 0.32s.

Contact Us
Book a free online demo or use the contact form to get in touch
Newsletter
Subscribe to our newsletter for regular updates

Select your desired system

The cutting edge test based outcome assessment system for health professionals and trainers

The most versatile toolkit for free data acquisition and comprehensive analytics in research

Have a general inquiry?

Write us a message for general questions about products and solutions or if you’d like to discuss other topics.


The form was sent successfully.

You will be contacted shortly.

moticon-rego-sensor-insole-live-event

Stay one step ahead!

Subscribe to our newsletter for the latest information on case studies, webinars, product updates and company news

Get support

Check our FAQ database for answers to frequently asked questions

Describe your issue in as much detail as possible. Include screenshots or files if applicable.


Need help?
Want a live demo?
Interested in prices?
Want to say hello?
Always just a call away
+49 89 2000 301 60