BrainGait: Gait Event Detection and Visualization for Robotic Rehabilitation

Information

• Publication Type: Master Thesis
• Workgroup(s)/Project(s):
  • Visualization Group
• Date: April 2020
• Date (Start): March 2019
• Date (End): 21. April 2020
• Diploma Examination: May 2020
• Open Access: yes
• First Supervisor: Eduard Gröller
• Keywords: Mobility rehabilitation, Electroencephalography

Abstract

Mobility impairment in adults is one of most prevalent types of disabilities in developed countries. Gait rehabilitation can be used to regain some or all motor functions, especially after a stroke. In recent years, robot-assisted gait training attracted increasing interest in rehabilitation facilities and scientific research. With this advent of robotic recovery comes the need to objectively measure the patient’s performance. Physiotherapists need essential information about the current status during training and how to improve the patient’s gait, presented in an easy to grasp and compact form. On the other hand, physicians rely on statistical measures in order to evaluate the patient’s progress throughout the therapy. This thesis discusses commonly used visualizations and statistics while proposing improvements and adaptations in the context of PerPedes, a novel robotic gait rehabilitation device. In order to measure the patient’s performance, a new algorithm for gait event detection was developed, based on force data from pressure plates. The following work demonstrates that standard algorithms fail with PerPedes, while the proposed solution can robustly handle highly distorted gait patterns, such as hemiplegic gait, foot drop, or walking backwards. The software application developed during this thesis provides feedback to the therapist and generates suggestions for gait improvement. Furthermore, gait statistics are inferred from each therapy session and collected in order to be used for future analysis and inter-patient comparison.

Additional Files and Images

Weblinks

• Entry in reposiTUm (TU Wien Publication Database)
• DOI: 10.34726/hss.2020.57203

BibTeX

@mastersthesis{Spelitz2020,
  title =      "BrainGait: Gait Event Detection and Visualization for
               Robotic Rehabilitation",
  author =     "Stefan Spelitz",
  year =       "2020",
  abstract =   "Mobility impairment in adults is one of most prevalent types
               of disabilities in developed countries. Gait rehabilitation
               can be used to regain some or all motor functions,
               especially after a stroke. In recent years, robot-assisted
               gait training attracted increasing interest in
               rehabilitation facilities and scientific research. With this
               advent of robotic recovery comes the need to objectively
               measure the patient’s performance. Physiotherapists need
               essential information about the current status during
               training and how to improve the patient’s gait, presented
               in an easy to grasp and compact form. On the other hand,
               physicians rely on statistical measures in order to evaluate
               the patient’s progress throughout the therapy. This thesis
               discusses commonly used visualizations and statistics while
               proposing improvements and adaptations in the context of
               PerPedes, a novel robotic gait rehabilitation device. In
               order to measure the patient’s performance, a new
               algorithm for gait event detection was developed, based on
               force data from pressure plates. The following work
               demonstrates that standard algorithms fail with PerPedes,
               while the proposed solution can robustly handle highly
               distorted gait patterns, such as hemiplegic gait, foot drop,
               or walking backwards. The software application developed
               during this thesis provides feedback to the therapist and
               generates suggestions for gait improvement. Furthermore,
               gait statistics are inferred from each therapy session and
               collected in order to be used for future analysis and
               inter-patient comparison.",
  month =      apr,
  address =    "Favoritenstrasse 9-11/E193-02, A-1040 Vienna, Austria",
  school =     "Research Unit of Computer Graphics, Institute of Visual
               Computing and Human-Centered Technology, Faculty of
               Informatics, TU Wien",
  keywords =   "Mobility rehabilitation, Electroencephalography",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2020/Spelitz2020/",
}