Design of an integrated upper extremity exercise tracking and analysis platform

• Main Author: Ngo, Wei Kang.
• Other Authors: Yeo Song Huat
• Format: Final Year Project
• Language: English
• Published: 2013
• Subjects: DRNTU::Engineering::Mechanical engineering::Bio-mechatronics
• Online Access: http://hdl.handle.net/10356/53568
• Institution: Nanyang Technological University

In today’s world, many developed countries are facing a critical issue with the shortage of manpower in the healthcare industry due to a growing and ageing population. One possible solution to help alleviate this problem is to optimize the way current rehabilitations is being carried out. Currently, these rehabilitations are labour-intensive as therapists are required to be present throughout the process to monitor their patient’s progress. As these processes are usually repetitive, technology can be used to help therapists keep track of them and thus free up valuable time for the therapists to carry out more consultations and assessments. Currently many rehabilitation systems leveraging on technology are expensive and not user-friendly. Thus there is a gap for a system which is relatively cheaper and user-friendly such that more patients can benefit from it. In addition to helping the patients, the system should also aim to reduce the therapist’s workload on repetitive tasks, such as rehabilitation exercises. In this way, the therapist can leave the patients to carry out the repetitive exercises on their own while they spend more quality time on consultation and assessment. The therapist will then be able to come back to check on the progress of the patient once the experiment is completed. In this project, a system consisting of 2 Inertia Motion Units (IMUs) to capture the upper extremity motion and 2 programs for converting and presenting of useful information for the therapists is developed. Experiments with healthy subjects doing the shoulder flexion exercise to test the system are carried out. From the experiment data, useful information such as the angles of motion of the upper arm and elbow, the number of repetitions and the number of correct repetitions where the subject achieved the target motion is computed. This information is then displayed with the developed program using charts for the angles representation and a stickman for the simulation of the motion. For accuracy testing of the system, the data recorded by the system are compared to visual analysis data using the videos taken of the experiments. Overall, the system is accurate to about 10o in angles calculation relative to visual analysis and with 97.5% accuracy in counting the repetitions. Last but not least, this system is flexible, versatile and can be easily extended to include the tracking of other types of exercises by adding additional exercise-specific algorithms to the system’s database.