Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Using generated propulsion to modulate robotic gait training
In this master project we want to explore whether the concept of propulsion as gait performance assessment and as explicit training goal is also applicable in rehabilitation robots, and how the robots and the concept can best be made compatible.
Gait rehabilitation robots such as the Lokomat from Hocoma (Switzerland) have been around for almost 20 years now, and are being used in research as well as clinical training of patients groups such as patients with spinal cord injury (SCI) and cerebrovascular accident (CVA).
Nevertheless, it is believed that the potential of such technology is still much greater than the currently clinically demonstrated benefit, especially in patient groups with very individually specific profiles, such as patients with CVA. The currently limited effectiveness has a number of causes, including the limited implementation of the principles of motor learning in the robot-supported training, a limited adjustment of training content to the needs of the individual patient, and in general the limited understanding of how to best train an individual patient.
A recent trend in rehabilitation research is an increasing focus on the forward propulsion a patient generates through each leg during gait. This is a quantitative metric that summarizes the effectivity of the efforts a patient is performing to move forwards, and is clearly distinguishable between paretic and non-paretic leg contribution in CVA. It has been demonstrated that providing bio-feedback and motivating patients to improve paretic leg propulsion, improves overall gait performance, and that the metric is strongly related to overall gait speed, which is a key assessment metric of impairment and recovery.
In these studies, propulsion was measured through an instrumented split-belt treadmill, where the patient is walking on own effort, with at most a level of Body Weight Support as assistance.
Gait rehabilitation robots such as the Lokomat from Hocoma (Switzerland) have been around for almost 20 years now, and are being used in research as well as clinical training of patients groups such as patients with spinal cord injury (SCI) and cerebrovascular accident (CVA). Nevertheless, it is believed that the potential of such technology is still much greater than the currently clinically demonstrated benefit, especially in patient groups with very individually specific profiles, such as patients with CVA. The currently limited effectiveness has a number of causes, including the limited implementation of the principles of motor learning in the robot-supported training, a limited adjustment of training content to the needs of the individual patient, and in general the limited understanding of how to best train an individual patient. A recent trend in rehabilitation research is an increasing focus on the forward propulsion a patient generates through each leg during gait. This is a quantitative metric that summarizes the effectivity of the efforts a patient is performing to move forwards, and is clearly distinguishable between paretic and non-paretic leg contribution in CVA. It has been demonstrated that providing bio-feedback and motivating patients to improve paretic leg propulsion, improves overall gait performance, and that the metric is strongly related to overall gait speed, which is a key assessment metric of impairment and recovery. In these studies, propulsion was measured through an instrumented split-belt treadmill, where the patient is walking on own effort, with at most a level of Body Weight Support as assistance.
There will be some room to adjust precise goals to individual interests and background of the student, among the following possible goals:
1a. Develop a method to measure generated propulsion in a robotic configuration, considering that the horizontal ground reaction force cannot be directly measured, so that the information has to be obtained from related data, such as forward push at pelvis level, trailing leg angle, or features of the center of pressure (several concepts are known from literature).
1b. Develop concepts for improving this measurement inside a robot configuration, i.e. adapting robot control, by combining available data, considering novel metrics or adding specific sensors
2. Evaluate how well a gait pattern with asymmetric propulsion can be quantified in a specific gait rehabilitation robot
3. Explore how the propulsion metric can be developed into an assessment metric that provides clinically relevant information on impairment and recovery of an individual patient
4. Explore how the propulsion metric can be used to provide bio-feedback, for example through using gamification concepts, develop and evaluate a prototype.
There will be some room to adjust precise goals to individual interests and background of the student, among the following possible goals:
1a. Develop a method to measure generated propulsion in a robotic configuration, considering that the horizontal ground reaction force cannot be directly measured, so that the information has to be obtained from related data, such as forward push at pelvis level, trailing leg angle, or features of the center of pressure (several concepts are known from literature). 1b. Develop concepts for improving this measurement inside a robot configuration, i.e. adapting robot control, by combining available data, considering novel metrics or adding specific sensors
2. Evaluate how well a gait pattern with asymmetric propulsion can be quantified in a specific gait rehabilitation robot
3. Explore how the propulsion metric can be developed into an assessment metric that provides clinically relevant information on impairment and recovery of an individual patient
4. Explore how the propulsion metric can be used to provide bio-feedback, for example through using gamification concepts, develop and evaluate a prototype.
Not specified
Enrolled in one of the following masters:
- Human Health Sciences and Technology
- Biomedical Engineering
- Mechanical Engineering
Good understanding of the biomechanics of walking, human motor control, and related sensing technologies, interest in human subject experiments, Matlab/Simulink- development skills
Potentially: skills to develop a game concept in Unity or similar environment
Enrolled in one of the following masters: - Human Health Sciences and Technology - Biomedical Engineering - Mechanical Engineering
Good understanding of the biomechanics of walking, human motor control, and related sensing technologies, interest in human subject experiments, Matlab/Simulink- development skills Potentially: skills to develop a game concept in Unity or similar environment
Peter Wolf, SMS lab
Jan Veneman, Hocoma AG, Switzerland – use of a Lokomat prototype at Hocoma premises in Volketswil will be possible and is expected in the context of this project.
If you are interested in the project please apply with your CV, academic transcript, and a short description of your interest in the project to:
jan.veneman@hocoma.com
Peter Wolf, SMS lab Jan Veneman, Hocoma AG, Switzerland – use of a Lokomat prototype at Hocoma premises in Volketswil will be possible and is expected in the context of this project.
If you are interested in the project please apply with your CV, academic transcript, and a short description of your interest in the project to: jan.veneman@hocoma.com