Robotic arm reaches toward success

A recent segment on 60 Minutes detailed the successes of the Hopkins Applied Physics Laboratory’s Revolutionizing Prosthetics team in mind-controlled prosthetics, also called Modular Prosthetic Limbs (MPL). Geoff Ling, DARPA program manager for the Revolutionizing Prosthetics program, answered some questions about the team’s work for The News-Letter.

The News-Letter (N-L): How long might it take to expand this research and then make an actual FDA-approved product from it that could be produced on a larger scale?

Geoff Ling (GL): The Johns Hopkins University Applied Physics Lab (APL) is one of two prime performers on DARPA’s Revolutionizing Prosthetics program; the other is DEKA Research and Development Corp. Both performers have developed prototype modular prosthetic arm systems, including sockets, which offer increased range of motion, dexterity, and control options to users.

From the beginning of the program, one of DARPA’s goals has been regulatory approval by the U.S. Food and Drug Administration (FDA) of the devices and control schemes created. In seeking regulatory approval, a device and its control scheme must be tested and approved together after demonstrating their safety and efficacy. Therefore, all clinical and take-home studies require the paired use of the arm and control scheme that will ultimately be considered for commercial transition.

At the start of the program, DARPA made team assignments to use invasive or non-invasive controls; “invasive” means requiring surgical implantation of devices. The intent was that DEKA could quickly get to market a mechanically advanced, anthropomorphic prosthesis with non-invasive controls, and that APL could, over the longer term, unite advances in neuroscience to develop devices that enable near-natural control of advanced arm systems through brain control of the prosthesis, possibly including sensory feedback to the brain.

There are many variables regarding timing. FDA approval of a brain-control scheme for an advanced prosthetic limb depends on the outcome of trials and cannot be predicted, nor can the scale of any production. Definition of the brain-controlled system must be achieved before it can be submitted to the FDA for approval, and some of the technology, the implanted array in particular, must go through another evolution of design before it can be part of an objective system.

Please note, though, that both performers’ arm systems can, in fact, be operated with the same control schemes, ranging from surface electromyography (EMG) to implanted arrays in peripheral nerves or the brain. Both arms use a CANBUS communication interface to facilitate standardizing input commands.

N-L: What were the biggest challenges in developing this technology (engineering or otherwise)?

GL: The entire Revolutionizing Prosthetics effort to date has required significant advances in engineering, but two major advances remain to be tackled before neural control of a prosthetic limb is viable over the long term. First, it will be necessary to improve device reliability, especially for implantable devices such as sensors that require surgery to access. Second, users need to be able to control the arm system wirelessly. Together, these developments would minimize the need for surgery and eliminate the need for the wires that control signals to pass through the skin or skull.

N-L: Are there plans to involve more quadriplegics in this research? Do you expect the technology to work with any person?

GL: Yes, one of APL’s subs, the California Institute of Technology, has begun recruiting for a quadriplegic research participant.

Yes, the technology can work with any person, but those who can will likely opt for the least invasive control technique possible. To clarify, the arm systems developed under Revolutionizing Prosthetics can be controlled with a variety of inputs, including: targeted muscle reinnervation, inertial measurement units outside the user’s body and surface electromyography. In the case of partial amputees, who have nerves still intact, direct brain control of an advanced prosthesis is possible by attaching existing nerves to the device. Again, the FDA will have to approve each arm-control modality before it can be made available.

For now the testing with neural implants has involved only quadriplegics due to a personal cost-benefit analysis by the patients regarding potential quality of life improvements versus the invasive nature of the required surgery, but it is by no means limited to quadriplegics.

N-L: You’ve had success with haptic feedback with amputees allowing them to regain some sense of touch (or is haptic the wrong term to use in this situation?).  How do you plan to adapt this success in the case of a quadriplegic so that they too could begin to feel their surroundings again?

GL: A goal of the program is to provide near-natural feedback through direct stimulation to the brain. This is in the experimental phase and announcements will be made if it is approved for human trials. To protect human patient privacy, announcements are controlled very carefully and only made with permission of the patient.