Home // ACHI 2021, The Fourteenth International Conference on Advances in Computer-Human Interactions // View article

Assessment of Drug Picking Activity using RGB-D Camera

Authors:
Yuta Ono
Oky Dicky Ardiansyah Prima

Keywords: Medication administration error, 3D human pose estimation, MediaPipe, RGB-D camera, Azure Kinect

Abstract:
Non-pharmacists have been allowed to pick drugs under the responsibility of pharmacists in order to reduce the burden on them in Japan. However, the activity tends to occur human errors since the name or shape of drugs are similar. While Bar-Code Medication Administration (BCMA) system and Automated Dispensing System (ADS) have been proposed to prevent such errors, these systems are cumbersome and costly. With the progress on human pose estimation technique using machine vision-based approach, it has become possible to measure the displacement and the posture of human body in real space. This approach makes us easy to measure the location of humans and other objects simultaneously. This study attempts to construct a drug picking activity judgement framework using RGB-D camera to detect the error easily at low-cost. This framework uses RGB-D camera to measure the location of hand landmarks and judges the activity by proposed judgement algorithm based on the position of those landmarks. In order to measure both hands accurately, we used Azure Kinect Body Tracking SDK and MediaPipe. Our experiments show that proposed framework is capable of the activity judgement on drug picking.

Pages: 6 to 11

Copyright: Copyright (c) IARIA, 2021

Publication date: July 18, 2021

Published in: conference

ISSN: 2308-4138

ISBN: 978-1-61208-870-9

Location: Nice, France

Dates: from July 18, 2021 to July 22, 2021