Home // eTELEMED 2020, The Twelfth International Conference on eHealth, Telemedicine, and Social Medicine // View article

Multi-action Detection System Using Infrared Omnidirectional Cameras

Authors:
Takashi Imabuchi
Yoshitoshi Murata
Oky Dicky Ardiansyah Prima

Keywords: hospital, patient monitoring, healthcare facility, ommnidirectional camera, human pose estimation

Abstract:
Japan is facing a shortage of healthcare workers due to the declining birthrate and aging population. This issue is placing a heavy burden on them to address the growing demand for 24-hour medical services. Information and Communication Technology (ICT) has opened up the possibility of collecting valuable data and providing insights for more control over patients' lives. Early approaches to detect accidental falling and wandering behavior using ICT include the use of invasive and non-invasive sensors. However, in order to put these approaches into practical use, further measures are needed. In this study, we propose a camera monitoring system to automatically detect typical patients’ behaviors, such as, rising from the bed, leaving the bed, falling down, and wandering. Our system utilizes infrared omnidirectional cameras that allows a wide range of monitoring during day and night. Skeletal information of multiple patients is captured using a computer vision-based pose detection to classify each behavior. Evaluation experiments demonstrated the feasibility of detecting typical patients’ behavior using the proposed system.

Pages: 30 to 35

Copyright: Copyright (c) IARIA, 2020

Publication date: March 22, 2020

Published in: conference

ISSN: 2308-4359

ISBN: 978-1-61208-763-4

Location: Valencia, Spain

Dates: from November 21, 2020 to November 25, 2020