Automatic Physiotherapy and Rehabilitation System for Lower Body Total Joint Replacement

Physiotherapists spend a large amount of their time with patients observing their rehabilitation techniques. A patient going through rehabilitation exercises are routine and does not necessarily require the attention of the physiotherapist. This research will develop a sensor system that will be strapped onto the patients and will provide feedback on how accurately the exercise is being executed. This will free up the physiotherapist to focus on diagnosis and other tasks that will better utilize the physiotherapist's training. A previous study has shown that this system is feasible for healthy subjects. This study would test to see if this system is extendable to rehabilitation subjects.

Study Overview

• Status: Completed
• Conditions: Lower-body Total Joint Replacement

Detailed Description

Many of the tasks performed regularly by physiotherapists during any given rehabilitation session are repetitive and do not rely on the physiotherapist's expertise, and could be performed and observed by automated means. The developed system will detect patient body postures and movements with data collected through sensors such as accelerometers. This data will be pattern matched to a predetermined movement pattern and feedback will be provided for patients regarding accuracy of their exercises. This data can be logged for the physiotherapist to examine at a later time. By automating this component of a rehabilitation session, the system will allow the physiotherapist to focus diagnosis and other tasks that will better utilize their training. The specific application for this prototype will be post-operative knee/hip replacement patients, so all devices must be non-invasive and must not interfere with normal recovery processes.

A previous version of this experiment on healthy participants has been successfully performed. This study would like to examine the feasibility of this system on rehabilitation subjects, as the movement patterns of a subject in physical rehabilitation may be dramatically different then a healthy subject. No intervention is suggested by the system, as this study is observational.

• Study Type: Observational
• Enrollment (Actual): 18

Contacts and Locations

Study Locations

• Canada
  • Ontario
    • Toronto, Ontario, Canada, M5G 2A2
      • Toronto Rehabilitation Instititue

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: No
• Genders Eligible for Study: All

• Sampling Method: Non-Probability Sample

Study Population

Specific target audience for sensor deployment is patients who has undergone knee or hip joint replacement and is taking physiotherapy rehabilitation sessions to recover. The target audience of this system is rehabilitation subjects who have recently had knee or hip replacement therapy. We would like to track several subjects over their rehabilitation stay to see the impact of rehabilitation on their movement pattern. Recruitment will be limited to in-patients only, as out-patients tend to be discharged from rehabilitation because they are healthier, and would likely have movement patterns closer to healthy subjects.

Description

Inclusion Criteria:

• Any patient who has had hip and/or knee total joint replacement and requires rehabilitation, and are assessed by physiotherapists as likely able to finish their rehabilitation cycle and be discharged to home.
• Patients with medical complications or other injuries will be not be excluded, as we are interested in seeing the movement profiles of a wide range of people.
• In-patients

Exclusion Criteria:

• Patients at risk of developing serious postoperative complications, such as an infection, myocardial infarction or anything that requires subsequent surgery, they may be excluded.
• Out-patients.
• Patients who cannot give explicit consent or understand the physiotherapist's instructions
• Patients who do not speak fluent English

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort
Total joint replacement; Patients who has undergone lower body total joint replacement surgery.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Joint angle assessment	Inertial measurement unit (IMU) data will be translated to joint angles via extended Kalman filter and kinematic modeling.	2 weeks

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Motion segmentation and identification	The joint angle data will be processed to automatically segment and identify rehabilitation motion.	2 weeks

Collaborators and Investigators

• Sponsor: University of Waterloo
• Collaborators: Toronto Rehabilitation Institute; Natural Sciences and Engineering Research Council, Canada
• Investigators: Principal Investigator: Dana Kulic, PhD, University of Waterloo

Publications and helpful links

General Publications

• Lin JF, Kulic D. Human pose recovery using wireless inertial measurement units. Physiol Meas. 2012 Dec;33(12):2099-115. doi: 10.1088/0967-3334/33/12/2099. Epub 2012 Nov 23.
• Feng-Shun Lin J, Kulic D. Segmenting human motion for automated rehabilitation exercise analysis. Annu Int Conf IEEE Eng Med Biol Soc. 2012;2012:2881-4. doi: 10.1109/EMBC.2012.6346565.
• Lin JF, Kulic D. Online Segmentation of Human Motion for Automated Rehabilitation Exercise Analysis. IEEE Trans Neural Syst Rehabil Eng. 2014 Jan;22(1):168-80. doi: 10.1109/TNSRE.2013.2259640. Epub 2013 May 2.

Study record dates

Study Major Dates

• Study Start: November 1, 2012
• Primary Completion (Actual): July 1, 2013
• Study Completion (Actual): December 1, 2015

Study Registration Dates

• First Submitted: December 3, 2014
• First Submitted That Met QC Criteria: December 18, 2014
• First Posted (Estimate): December 24, 2014

Study Record Updates

• Last Update Posted (Estimate): June 20, 2016
• Last Update Submitted That Met QC Criteria: June 16, 2016
• Last Verified: June 1, 2016

More Information

Terms related to this study

• Keywords: Physiotherapy; Total joint replacement; Kalman filtering; Kinematic modeling; Hidden Markov model; Motion segmentation
• Other Study ID Numbers: ORE 18193; REB 12-025 (Other Identifier: University Health Network)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO