Sensor Based Feedback

        Exercise is a powerful intervention in rehabilitation and prevention of disability. With outpatient physical therapy sessions typically costing over $100, correct adherence to supplemental home exercises is essential for less expensive, safe, and effective care. However, it is difficult to achieve patient adherence to a home exercise program  it requires a great deal of practice (thousands of repetitions) to learn new movement patterns  and incorrectly performed exercises are ineffective or even dangerous.  For these reasons, many prescribed therapeutic exercise programs require continuous supervision by a clinician    

       We  are developing a system to address the need for accurate, consistent and safe home exercise performance. By merging wearable inertial sensors and a  Kinect device  this system can capture, record, and process the exerciser’s movement while concurrently providing targeted feedback to guide correct exercise completion. A clinician programs the system with the exercise(s), including relevant parameters for each, similar to the tuning of exercise performance that the clinician does in person in the clinic. This new technology has the potential to decrease health care costs and increase access to care while improving the quality of that care.      

        This pilot study allowed us to perform initial testing of the system.  Specifically, we examined the effectiveness of the Kinect camera feedback in improving exercise accuracy, with and without the addition of inertial sensors, as compared to a traditional written home exercise program.  This work has been performed in collaboration with Dr. Wenbing Zhao and Dr.  Nigamanth Sridhar (CSU College of Engineering).  This work has been supported, in part, by a Faculty Research Development grant from CSU and by a Summer Undergraduate Research Award.  It is currently supported, in part, by a grant from the Ohio Bureau of Workers Compensation.  Please see below.  


Pictures: Top and Middle) Early prototypes of body worn, accelerometer based sensors. Bottom) Feedback for correct exercise performance using kinect based sensor technology.  

Publications: 

Zhao W, Lun R, Gordon C, Fofana A, Espy D, Reinthal A, Ekelman B, goodman G, Niederriter J, Luo C, Luo X.  Lifting

Done Right: A Privacy-Aware Human Motion Tracking System for Healthcare Professionals. Special Issue on 2016 IEEE

International Conference on Electro/Information Technology (EIT 2016).  Volume 7 • Issue 3 • July-September 2016

Zhao W, Lun R, Gordon C, Fofana A, Espy D, Reinthal A, Ekelman B, Goodman G, Niederriter J, Luo X. A Human

Centered Activity Tracking System: Towards a Healthier Workplace.  IEEE Transactions on Human-Machine Systems.

Zhao W, Lun R, Gordon C, Fofana A, Espy D, Reinthal A, Ekelman B, Goodman G, Niederriter J, Luo X. A Privacy

Aware Kinect-Based System for Healthcare Professional. IEEE International Conference on Electro Information 

Technology, Grand Fork, ND, pp. 205-210, May 19-21, 2016.

Zhao, W., Goodman, G., Ekelman, B., Espy, D., Niederriter, J., & Reinthal, M. (2015). An Integrated System for 

Privacy-Aware Human Motion Tracking with Realtime Haptic Feedback. In IEEE MS 2015. New York

W. Zhao, R. Lun, D. Espy, and M. A. Reinthal.  Realtime Motion Assessment For Rehabilitation Exercises: 

Integration Of Kinematic Modeling With Fuzzy Inference.   JAISCR, 2014, Vol. 4, No. 4, pp. 267 - 285.   Journal 

of Artificial Intelligence and Soft Computing Research

W. Zhao, D. Espy, M. A. Reinthal, and H. Feng, A Feasibility Study of Using a Single Kinect Sensor for 

Rehabilitation Exercises Monitoring: A Rule Based Approach, in Proceedings of the IEEE Symposium on 

Computational Intelligence in Healthcare and e-Health, Orlando, Florida, USA, December 9-12, 2014.

W. Zhao, R. Lun, D. Espy, and M. A. Reinthal, Rule Based Realtime Motion Assessment for Rehabilitation 

Exercises, in Proceedings of the IEEE Symposium on Computational Intelligence in Healthcare and e-Health, 

Orlando, Florida, USA, December 9-12, 2014.

Zhao, W., Espy, D.D., Reinthal, M.A., & Hai, F.(2014). Feasibility Study of Using Microsoft Kinect for Physical 

Therapy Monitoring. Encyclopedia of Information Science and Technology, Third Edition. 5542 – 5554.

Zhao, W., Feng, H., Roanna, L., Espy, D.D., & Reinthal, M.A.(2014).A Kinect-Based Rehabilitation Exercise

Monitoring and Guidance System. 5th IEEE International Conference on Software Engineering and Service

Science, June, 2014. 762 – 765.

Presentations: 

Use of two forms of real-time visual feedback to improve exercise accuracy.  Combined Sections Meeting of the

APTA, Indianapolis, IN, February 2015.  Ann Reinthal presenter with Debbie Espy, Jeffrey Swiers, and Philip 

Simon. 

Funding:

Ohio Bureau of Workers Compensation/Ohio Occupational Safety and Health Research

Program.  “Safe Patient Handling among STNA’s in Nursing Homes: Compliance, Monitoring, and Continuous 

Quality Improvement of Best Practices” Submitted 12/8/2014.  Requested $246,462 for 2 years.  Co-PI, with 

Glenn Goodman (COSHP, Occupational Therapy Program) as Program Officer, and others from COSHP, 

College of Engineering, School of Nursing, and external partners as Co-PI’s, Contractors, and Practice Team. 

[including    [Espy D, Reinthal A, and Zhao as co-PI's].

Reinthal A, Espy D, Sridhar N. $11,190. Utilization of feedback from inertial vs. Kinect sensors in improving 

exercise accuracy. CSU 2014 Summer Undergraduate Research Award Program. March 2014.

Reinthal A, Sridhar N. $9060. Initial feasibility testing of a system to provide home exercise guidance. 2013 

Engaged Learning Proposals: CSU Undergraduate Research and Creative Achievement. April 2013.

Espy D, Reinthal A, Zhao W, Sridhar N. $18,236. Development of the Technology for Exercise Tutor: a System

for Proxy Exercise Guidance. CSU Faculty Research Development Grant, 2012

Zhao W, Reinthal A, Sridhar N. $13,154. A low cost motion analysis system based on the Kinect. 2012

Engaged Learning Proposals: CSU Undergraduate Research and Creative Achievement. April 2012.