IMU based activity detection for post mini-stroke healthcare

Yi Wen Cheng, Pei Yu Chen, Chan Yun Yang, Hooman Samani

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Based on an IMU motion sensitive device, an activity detection system has preliminarily developed for the use of post mini-stroke healthcare. The electronically assistive device seeks a solution to identify specifically different motions of the post stroke patients. The development causally regards to a machine learning skill. To manage the input compatibility to the learning machine, a preprocessor of discrete action pattern length equalizer has thus been requested. The study introduces two types of the length equalizer to embed with the post ANN classifier to carry out the machine learning for the activity detection. By a general comparative evaluation, the study evidentially analyzed and ensured an excellently combinational devise for the detector with an empirical cross-validation procedure.
Original languageEnglish
Title of host publication2016 IEEE International Conference on System Science and Engineering, ICSSE 2016
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
ISBN (Electronic)9781467389662
DOIs
Publication statusPublished - 24 Aug 2016
Event2016 IEEE International Conference on System Science and Engineering, ICSSE 2016 - Puli, Taiwan, Province of China
Duration: 7 Jul 20169 Jul 2016

Publication series

Name2016 IEEE International Conference on System Science and Engineering, ICSSE 2016

Conference

Conference2016 IEEE International Conference on System Science and Engineering, ICSSE 2016
Country/TerritoryTaiwan, Province of China
CityPuli
Period7/07/169/07/16

Fingerprint

Dive into the research topics of 'IMU based activity detection for post mini-stroke healthcare'. Together they form a unique fingerprint.

Cite this