Session: 05-12-02: Robotics, Rehabilitation - II

Paper Number: 95215

95215 - Testing of a Robotic Prosthetic Leg 

We report on our design and evaluation of a prototype robotic prosthetic leg (RPL) with a powered non-back drivable knee and a hydro-pneumatic passive-resistive articulated ankle on a person with a leg amputation. We developed this prototype device to provide powered knee motion in ambulation while decreasing ankle joint stiffness in stance, improving knee stability on level ground and during ramp descent, and optimizing transverse joint-axis alignment to enhance symmetry and efficiency in ambulation. Our previous work demonstrated the safe function of this prototype on a non-disabled male subject on both level and sloped ground.

Continuing our evaluation and feasibility testing of this device, we recruited a single (n=1) 28-year-old female subject with a right knee-disarticulation (KD) amputation. She is 162 cm tall, 76 kg, and has a Medicare Classification Functional Level (MCFL) of K3.

For this case study, we configured the RPL to match the subject's lower leg height (45cm) and shoe size (22 cm). The knee unit is powered by a brushless DC motor (Maxon EC40) with a Hall sensor and a two-channel optical incremental encoder. The custom-built gearbox with elliptical gearing provides a 360:1 reduction to meet the walking and standing power requirements of a K2 or K3 ambulator. The ankle incorporates three passive hydro-pneumatic dampers, which provide tunable plantar flexion resistance and dorsiflexion assistance. The ankle unit was mounted to a custom-built alignable carbon fiber foot and instrumented with two load cells. Both foot and ankle are instrumented with two inertial measurement units (IMUs). The subject uses a handheld human-machine interface (HMI) to manually position the knee and change the speed of knee motion when necessary. An automatic mode is provided to initiate a complete step with the push of a button. A 3D printed cover, derived from a scan of the contralateral limb is used as a protective covering.  The total mass of the RPL in this configuration is 4.11 kg. A secondary HMI is equipped with a touch screen, providing the researcher or clinician additional monitoring and control of the system and data collection.  The complete control system and battery supply are housed in a tethered backpack that weighs 5.3 kgs worn by a researcher.

A certified prosthetist tuned the RPL alignment to match the non-amputated leg of the subject. The subject demonstrated walking with the RPL at various cadences and velocities in the ten-meter walk test (10MWT) and the six-minute walk tests (6MWT). The subject achieved a walking velocity of 0.77 m/s, unsupported and without an assistive device. The device's kinematic and kinetic testing was conducted in a motion capture lab while the subject performed a series of walking tests with their physician-prescribed prosthesis (RxPx) and with the RPL. Kinematic data was recorded using a 24-camera motion capture system. The data comparison showed improved ankle and hip motion with the RPL as compared to the RxPx. 

We demonstrated the RPL to be safe and feasible in ambulation tasks used by a female with a KD amputation.  These results demonstrate that this prosthetic device operated safely in a controlled environment and is feasible for further testing among larger populations of persons with limb loss.

Presenting Author: Michael Davidson Loma Linda University O&P

Presenting Author Biography: Michael Davidson is a PhD candidate in Rehabilitation Sciences at Loma Linda University. He is also an Assistant Professor for the School of Allied Health Professions and School of Medicine. Additionally, Michael is the Clinical Manager of the the Orthotics & Prosthetics Department at Loma Linda University Health. He has a Master of Science in Bioengineering from University of California Riverside, a Master in Public Health from Loma Linda University and a Bachelor of Science in Orthotics & Prosthetics from Cal State Dominguez Hills. He is certified in orthotics & prosthetics.

Authors:

Michael Davidson Loma Linda University O&P
Noha Daher Loma Linda University, School of Allied Health Professions
Thomas Fryer University California, Riverside - Bourns College of Engineering
Robert Dudley Loma Linda University, School of Allied Health Professions
Johannes Schaepper Loma Linda University - School of Allied Health Professions
Duc Tran Loma Linda University - Department of Physical Medicine and Rehabilitation