Thought Controlled Artificial Limbs



I wrote about the possibility of brain-controlled artificial limbs in “The Future of Medicine” but now there has been real progress. At Johns Hopkins Applied Physics laboratory, scientists have progressed with their design of an artificial limb that will have a brain controlled interface. The model came about through a contract with the Defense Advanced Research Projects Agency (DARPA) which has been looking for a prosthetic arm that would be many leagues advanced from those in use today and which in fact date back to the World War II era.

Not all that much progress has been made over the past few decades in artificial arm development. Perhaps it is because losses of legs are much more common than losses of arms. But the loss of an arm is especially devastating to the individual and a truly useful replacement is of critical need.

The new device will have remarkable dexterity with the degrees of freedom of a human wrist and the ability to control individual fingers. Look at your wrist. It can move in six different directions or “degrees of motion.” When you consider the entire arm, there are 27 degrees of motion and the new limb will have about 22 of them included. It weighs just eight pounds which is about what an average arm weighs yet can hold up to fifty pounds. The motive power comes from a rechargeable battery. These are advances of some great import indeed but the next step is the amazing one – brain control. The first step is to use outputs from the nerves in the shoulder that used to control the arm before the injury and loss. These nerves carry outputs from the brain that can be accessed to drive the various internal motors that operate the artificial arm. Later, the plan is to develop microchips to implant in the brain that will sense the “thought” to, say, “lift the arm” or “push that button.”

Johns Hopkins APL is engaged with multiple other groups to bring this work to fruition. One of the major hurdles is to develop the algorithms that take the signals from the brain or the nerve and convert them into mechanical activity. Signal analysis algorithms have now been developed that take outputs from the motor and the premotor cortex of the brain and decode them into specific dexterous movements such as grasping that can drive the electro-mechanical apparatuses in the limb.

The research needed to move this project ahead are daunting but the principals believe that the technology exists and can be turned to good use here. Perhaps one of the first types of patients to be tested will be quadriplegics because to offer such an advance would be dramatic for the involved patient. It sounds like science fiction but instead it is the result of the combined efforts of many engineering and computer scientists along with rehabilitation physicians and others.