Advances in Prosthetic Hands

by Emmanuele Gruppioni, M. Eng.

The human hand has 21 degrees of freedom (DOFs); therefore, its loss is probably the most traumatic type of amputation, not only given the functional aspects carried out by the hands, but also in consideration of its roles in our social life (such as shaking hands, or using the hand to greet, for example).

The common treatment for a trans-radial or hand amputation is a prosthesis with just one DOF for the hand (one motor drive which opposes the thumb and index and medium fingers) and (eventually) one degree of freedom for the wrist, which is typically controlled by two electromiographic signals coming from the residual muscles of the stump (normally the muscles to flex or extend the wrist). This is a good solution for the reliability, the toughness, and the grasping power, but it has great limitations in the appearance and in performing fine activities (such as using a computer or a cell phone).

For a long time, research has been working on new robotic systems with the aim of overcoming the limitations of traditional tridigital myoelectric prosthetic hands and to decrease, in this way, the gap with the human hand. 

The poly-articulated prosthetic hands are the first answer.

In a few words, poly-articulation is a concept based on a simple consideration: the main distance between actual prostheses and the human joints is due to the number of DOFs, so, what is needed is to increase the number of DOFs.

Simple to say, but not so simple to do.

To increase the number of DOFs means to improve the functionality and the appearance but, on the other hand, also to put much more complexity into the prosthesis structure; it requires extreme miniaturization of the mechanisms and of the motors, and more sophisticated electronics and software. Each one of these issues is a challenge for technology because, in addition, the prosthetic solutions have to be lightweight, with high power grasp and human-like appearance; these specification are obviously in mutual collision.

The commercial poly-articulated hand prostheses mount five motors that drive the robotic fingers through a series of gearboxes that obtain the right velocity, a no-driveback behavior with the minimum loss of efficiency so to reach a good user power and save the battery.

Basically, there are two technical solutions for the polyarticulated hands: one with the motors/mechanisms inside the first phalanx of the fingers and the other one with the motors/mechanisms inside the palm.

The first one is the "expensive" solution that requires special motors and high-level machining of the gearboxes. The functional unit is the single finger; therefore, this device allows treatment not only of complete hand amputations, but also partial hand amputations.

The second one is the "cheaper" solution because it uses more economic, basic components, but, obviously, it is not possible for the the treatment of partial hand amputations. A significant difference is that this solution implements a position closed loop control, whereas the other is simply based on the currents supplied to the motors.

Both the devices are controlled as the tridigital hands (with only two control signals) but the users can put these hands in a series of predefined postures through special strategies and, from there, move the hand in different ways. For example, the electronic unit can be programmed to detect the co-contraction of the muscles and, in this case, to put the prosthesis automatically in "index point" posture (with the index finger totally extended and the other fingers totally flexed, commonly used for computer or cell phone use).

It's important to understand that 1) the patient is not able to control the single finger and 2) there is no correspondence between what the patient thinks/wants to move and the specific joint activation. In fact, it's impossible to obtain from a stump a large number of independent and strong muscle signals that could give to the patients a better control of this type of prosthesis. This is one of the main fields of research in prosthetics.

In conclusion, the first poly-articulated hands are now available on the market and the benefits for the patients are clearly visible. This is just the initial result of the research activity whose final goal is a much more functional, dexterous and human-like prosthetic hand. However, a lot of challenges are still to be faced, particularly from the control viewpoint.