This article was updated on the 12th September 2019.
Andrii Zastrozhnov / Shutterstock
A prosthetic limb is defined as a mechanical device that is used to replace a missing human limb. The device is designed to help the user coordinate better control of an amputated limb. This could be as a result of motor control loss by a traumatic event, a congenital-related defect, or dyvascular-related.
According to statistics by Ziegler-Graham, et al. (2008), an estimated 1.6 million civilians were living with the loss of a limb. The research also revealed that approximately 38% of these people suffered an amputation of a limb as a secondary consequence to a dyvascular disease. Shockingly, it has been predicted that this statistic is likely to double to 3.6 million by 2050.
Dyvascular disease is only one major contributing factor to the amputation of a limb. It is commonly known that war veterans often become wounded in combat and considering the current affairs, it is estimated that the number of veterans wounded in Afghanistan is increasing. According to the Department of Defense statistics, to-date more than 1000 soldiers have returned from Afghanistan with the loss of a limb. With current limb-loss statistics in perspective, there is a clear demand for the use of prosthetics from the healthcare angle.
Robotic Prosthetics - Structural Components
A robotic prosthetic limb is made up of the following components:
- Biosensors - these are required to detect neural transmission from the human nervous and neuromuscular junctions (i.e., wires to track neural activity near the surface of the skin). The video below by Bio Tac is a great example of how biosensors are applied to study sensing modalities of a robotic finger.
Overview of BioTac Sensory Technology
- Controller - the main connection between the neuromuscular systems and the end device and so this receives the information from the biosensors and feeds this information back to the actuator.
- Actuator - the end device that will receive information from the controller and mimic the movement of a muscle.
The study of prosthetic limbs is vast and complex. For the scope of this article, attention will focus on the prosthetic hand in more detail. As mentioned, the main pathway involved in the function of a prosthetic limb is based on three main components: a biosensor, controller, and actuator. User preferences need to be considered before designing a prosthetic limb and if focusing on a hand prosthesis, the following preferences are important:
- Grasp function for a size range of objects
- Intricate finger movements should be possible for grasping and pinching motions
- A prosthetic hand needs to be lightweight to allow for better movement in continuous space
- Finger projections need to be designed with active joints to allow for intricate movement
- Aesthetically, the prosthetic hand needs to appeal to the patient and allow for comfort of use
The use of biosensors is fundamental to initiating a pathway that will result in movement of the actuator. To allow for the control of a prosthetic limb, the actuators are attached to the residual part of an amputated area, which will provide feedback on tactile information generated by the biosensors. The actuators are also connected to a hardware interface system that acts as the controller, which initiates sensory feedback to the actuators.
The actuator is the key element to a fully functioning prosthetic hand and it is what displays the end result (i.e., grasping an object). To begin with, the actuator system to this prosthetic hand is made up of micro-actuators. These drive the metacarpo-phalangeal (MP) joints of the thumb and the proximal interphalangeal (PI) joints. The distal interphalangeal (DIP) joint is controlled by a link connected to the proximal interphalangeal (PIP) finger joint.
It is important to consider the number of individual pieces of information that will predict how a certain parameter will behave (degrees of freedom [DOF]). With this in mind, it is known that each finger to a hand has 4 DOF and the wrist has 3 DOF, which means that a prosthetic hand has to be designed to function with 23 DOF. It is quite difficult to control the movement of such a complex machine and so a basic two to three fingered prosthetic hand is usually designed for better control over the monitoring of signal patterns. The thumb actuator works with two degrees of freedom (DOF), which are comparative to the MP and IP joints. The modeling of heavy actuators results in a lower DOF, which can affect the ability of the prosthetic hand to grasp objects effectively. The following video is a good example of a patient adapting a myoelectric prosthetic hand controller with signal activity being monitored.
Emily Von Dollen - adapting her Myoelectric Prosthetic Hand Controller
To take this journey of prosthetic hand construction a little further into detail, we need to consider the application of the sensor-controller-actuator system when measuring micro-electric signals. An electromyography (EMG) signal is used to detect electrical stimulus from the robotic hand, as would have been the case if the real limb was present. Electromyography involves the use of surface electrodes to monitor muscle activity and when considering bionic signals, this method of signal control is ideal to monitor exactly how the mechanical device is moving and responding to stimuli. The EMG is particularly important as it provides the functionality to amputees and the basic principle of EMG signals is to pick up a signal based on the movement of a limb or robotic form. The video below demonstrates the use of a robotic hand with one degree of freedom to pick up an object and how EMG is used to map the electrical signals during this movement.
EMG-Controlled Prosthetic Limb
Research to Consider
The present article has introduced a basic overview of the design and modeling of a prosthetic hand, a study referred to as biomechatronic design. Though the design of prosthetics is continuing to develop and benefits many patients living with an amputated limb, there are still challenges ahead in the design of a prosthetic limb that satisfies intricate requirements, such as easy control of the prosthetic limb and a to make this mechanical device cosmetically appealing. There is also the challenge of understanding the issue of tissue reactions to material used for the prosthetic limb and how an inflammatory response to such a reaction may interfere with the signal transmission of biosensors.
A lot of work has gone into further understanding and replicating the movements of a human limb to better design robotic hands that can perform as well as or better than human hands. Analysis of available prosthetic hands using computational approaches, revealed that symmetrical designs preferred by roboticists or the random designs generated come close to the abilities of a human hand. Thus, prosthesis designed with anatomically plausible asymmetry in joints and tendons can perform even better than human hands.
Another approach is to replicate the biomechanics of a human hand using biomimetic, mechanical design. The hands can be controlled via methods such as using electromyography signals from the residual limbs or by implanting tiny electrodes in the motor region of the brain or nerves responsible for controlling the hand. To control the movement of the fingers, new approaches have used cable-driven systems where strings replicate the tendons of the hand rather than conventional mechanical joints and transmission systems. This enabled the easier control of fingers as information about movement could be directly passed on to the prosthetic, rather than using complicated control systems.
The latest designs combine machine learning and neural networks to enable a user to simply think of a movement and the prosthetic performs it, just like a natural limb. A recent study reported a prosthetic hand combined with a neural network that can allow the wearer to think and move the hand. Electrical signals from the nerves are measured through the skin and sent to a computer that decides what the movement should be. The computer learns what the movement should be using a neural network trained to recognize different movements.
Another aspect of research into prosthesis has been enabling a prosthetic limb to feel. One report used sensors in the fingers to send signals to the nerves of the limb via an array of microelectrodes and wires that translate into a sensation. They may also be used to feel pain and temperature. Another approach has been to use a layer of artificial skin with sensors that are placed over the fingers on a prosthetic hand. The sensors relay the stimuli back to the nerve endings of the limb. Such a system can be used with any type of prosthetic hand.
Although there have been significant developments in prosthetics in the last few years, there is still a need for cheaper and more accessible prosthetics. More prosthetics are moving out of lab-scale testing and into the real world, opening up new opportunities and challenges.
- Ziegler-Graham K, MacKenzie EJ, et al. Estimating the preference of limb loss in the United States: 2005 to 2050. Archives of Physical Medicine and Rehabilitation. 2008: 89(3):422–9.
- Pitkin, M.R. (2006). Biomechanics of Lower Limb Prosthetics. New York: Springer Science and Business Media.
- Chen, C. (2009). Hybrid Control Strategies for Smart Prosthetic Hand. Arbor, Michigan: ISTE Publishing.
- Love, L.J. (2009). Mesofluidic Actuation for Articulated Finger and Hand Prosthetics. The 2009 IEE/R S J International Conference on Intelligent Robots and Systems. OCTOBER 11-15, 2009 St. Louis USA.
- Nasser, S., Rincon, D., Rodriguez, M. Design of an Anthropomorphic Underactuated Hand Prosthesis with Passive-Adaptive Grasping Capabilities. 2006 Florida Conference on Recent Advances in Robotics, FCRAR 2006.
- Dr A.H.Bottomley and T.K. Cowell. An Artificial hand controlled by the nerves. New Scientist. 1964 (No. 382).
- Carrozza, M.C., Micera, B., Micera, S., Zecca, M., Dario, P. A “Wearable” Artificial Hand for Prosthetics and Humanoid Robotics Applications. Proceedings of the 2001 IEEE-RAS International Conference on Humanoid Robots.
- Lai, J.C.K., Schoen, M.P., Gracia, A.P, ET AL. Prosthetic devices: challenges and implications of robotic implants and biological interfaces. Journal of Engineering in Medicine. 2006: Vol.221 Part H.
- Antfolk, C., Balkenius, C., Lundborg, G., et al. Design and technical construction of a tactile display for sensory feedback in a hand prosthesis system. BioMedical Engineering Online. 2010, 9:50. 1–9.