Home // ACHI 2022, The Fifteenth International Conference on Advances in Computer-Human Interactions // View article

3D Human Pose Estimation using a Stereo Camera towards Monitoring of Drug Picking Tasks

Authors:
Yuta Ono
Oky Dicky Ardiansyah Prima

Keywords: 3D human pose estimation, Stereo camera, Medication dispensing error, Pharmacy

Abstract:
Medication dispensing errors are a critical issue that threatens the quality of health care services and patients’ safety. Checks by pharmacists are effective in preventing this potential problem but increase the workload. In Japan, dispensing assistants are allowed to assist pharmacists in picking drugs to reduce their workload. However, becoming a dispensing assistant requires experiences, as drugs can easily be mistaken because of their similar names and forms. This study attempts to construct a monitoring framework using stereo camera-based Three-Dimensional (3D) human pose estimation for detecting errors and evaluating the physical workload during the drug picking task. To accurately estimate the 3D human pose, we improve the estimation of body joints and perform calibration using multiple reference points. Our results show that the proposed framework can estimate the 3D human pose with acceptable accuracy to detect errors in the drug picking task. Future work will examine the applicability of the proposed framework to the assessment of physical workload of dispensing assistants.

Pages: 11 to 17

Copyright: Copyright (c) IARIA, 2022

Publication date: June 26, 2022

Published in: conference

ISSN: 2308-4138

ISBN: 978-1-61208-982-9

Location: Porto, Portugal

Dates: from June 26, 2022 to June 30, 2022