Extracts key metrics from motion tracking data (in MVNX / XML format). Extract data can be used to determine if lifting and transfer movements risk injury. Software is based on research paper presented at ACM Digital Health Conference (DigH) 2017.
Builds under Python v2.7
By default data is input in the same source directory as source code. Four sample data files are provided. However, by default, the code runs a single file (0_Bed_to_ShowerChair_M.mvnx). You can add the others .mvnx files (or your own data) by uncommenting/editing main.py. If you input MVNX or XML file follows different formatting rules, you will need to edit MotionCaptureExecutor.py and edit the parsing rules.
By default results are placed in the results.csv
Please read the following research paper for details. If you have any questions, please feel free to contact me.
Users are encouraged to use, modify and extend this work under the GNU GPLv3 license.
Please cite the following paper to provide credit to this work:
Jonathan Muckell, Yuchi Young, and Mitch Leventhal. 2017. A Wearable Motion Tracking System to Reduce Direct Care Worker Injuries: An Exploratory Study. In Proceedings of DH ’17, London, United Kingdom, July 02-05, 2017, 5 pages. DOI: http://dx.doi.org/10.1145/3079452.3079493