Enhancing Predictive Power of Cluster-Boosted Regression with Text-Based Indexing

Journal article

Authors/Editors

Strategic Research Themes

No matching items found.

Publication Details

Author list: Kongburan W., Chignell M., Charoenkitkarn N., Chan J.H.

Publisher: Faculty of Engineering

Publication year: 2019

Journal: Engineering Journal (0125-8281)

Volume number: 7

Start page: 43394

End page: 43405

Number of pages: 12

ISSN: 0125-8281

eISSN: 0125-8281

URL: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85065075973&doi=10.4186%2fej.2019.23.2.71&partnerID=40&md5=4d5ccbaf753b3691c402aa8198b41b06

Languages: English-Great Britain (EN-GB)

View in Web of Science | View on publisher site | View citing articles in Web of Science

Abstract

The chest expansion measurement is a part of physical therapy to track the progress of rehabilitation for checking the performance of lungs. The chest expansion mechanism moves like pump handle and bucket handle. These two movements cause the chest move many directions. For tracking the chest movement in each direction, the MEMS accelerometers are used to measure acceleration in each axis. Consequently, acceleration is converted to displacement by double integration. The acceleration from accelerometer in each axis is affected from the earth gravity force. Thus, rotation matrix is used for compensating the earth gravity force. It can track the acceleration vectors while rotating. The known movement from robot is simulated similar the chest expansion. 60 sets of accelerometer data were collected from robot demonstration and were analyzed for testing the accuracy of sensor and algorithm. For the highest expansion, the chest expansion measurement must be performed while doing deep breathe inhale and exhale. The deep breath signal is a low frequency and there is high frequency noise. Therefore, a low-pass filter was used for eliminating high frequency noise. The accelerometers and VICON's markers were placed together on the body. The displacement results from accelerometers were compared with the displacement of VICON motion analysis system to find the accuracy of our purposed device. The average error of 20 sets of acceleration data from accelerometers which referred with VICON motion analysis is 7.195ฑ4.361 mm. Accelerometer result trend follows VICON motion analysis and corresponds to each other. ฉ 2019, Chulalongkorn University. All rights reserved.

Keywords

Chest expansion, Displacement, Gravity compensation, Rotation matrix