Proprioception is the sense of the position, speed, and torque of one’s body parts. In other words, it is what allows us to know where our body parts are as they move, without looking at them. It is fundamentally distinct from the sense of touch, although equally important.
Proprioception is caused by contraction and stretching of opposing muscles that form an agonist-antagonist pair. When you flex your ankle, the muscles in the front of your leg contract, causing the joint to move, and simultaneously stretching the muscles in the back of your leg. When you extend your ankle, the opposite happens; the muscles in the back of your leg contract, moving the joint and stretching the muscles in the front. When these muscles contract and stretch, biological sensors within the muscles pass information through the nerves to the brain, telling the brain that the joint is moving. This dynamic relationship between agonist and antagonist muscles is essential to our ability to sense where our limbs are in space.
The sense of joint position, speed, and torque is essential to all human movement. It is what allows us to reach for a coffee cup without knocking it over, or catch ourselves when we trip unexpectedly. Without proprioception, it is impossible to precisely control the movement of a limb.
The AMI is a method to provide proprioception from a synthetic device to the human nervous system. An AMI is made up of two muscles – an agonist and an antagonist – mechanically connected so that when the agonist contracts, the antagonist is stretched, and vice versa. The purpose of an AMI is to control and interpret proprioceptive feedback from a bionic joint. During an amputation procedure, a surgeon creates AMIs by linking together muscle pairs within the amputated residuum. Multiple AMI muscle pairs can be created for the control and sensation of multiple prosthetic joints.
When the AMI patient wishes to move his bionic limb, he contracts the AMI muscles associated with his intended joint motion. Muscle electrodes adjacent to the AMI muscles send electrical signals from the muscles to small computers on the prosthetic limb, which then use the muscle signals to control motion of the prosthetic joints in a natural way. Because the AMI agonist and antagonist muscles are mechanically connected within the residual limb, contraction of the agonist causes stretch in the antagonist. This coupled movement enables natural biological sensors within the muscle-tendon to transmit electrical signals to the central nervous system, communicating muscle length, speed, and force information, which is interpreted by the brain as natural joint proprioception.
Most neural interfaces are designed to send movement commands from the nerves to a prosthetic device, but do not provide feedback from the prosthesis back to the nervous system. Those systems that do provide feedback are limited to touch sensation, but are not able to restore proprioception in a robust, repeatable manner. They rely on implanted synthetic sensors that are held in direct contact with nerve tissue, and both send electrical impulses to and record electrical signals from the nerve. There are two key issues with this type of interface. First, the electrical signals that muscles use to communicate proprioception to the brain are quite complex, and it is not yet possible to reproduce these signals using artificial electrical stimulus. Second, nerves respond negatively when they are held in contact with synthetic materials. These two hurdles have made it difficult for neural technologists to create systems that are able to provide proprioception from a prosthetic device.
In contrast to these more conventional approaches, the AMI takes advantage of natural biological sensors within muscle and tendon to provide proprioception to the central nervous system. With this approach, rather than needing to speak the electrical language of nerves ourselves, we can rely on the biological sensors in the AMI muscles to translate mechanical stretch into electrical signals that can be interpreted by the brain as sensations of joint position, speed and torque. In addition, because the AMI serves as an intermediary with the nerve, it is no longer necessary that any synthetic material touches the nerve directly. In this way, the AMI is both more effective and more viable than an approach that relies on direct neural stimulation.
The AMI reaches its full potential as an interface to control and interpret proprioceptive feedback from a bionic joint. In a study published in Science Translational Medicine, we demonstrated that, when compared to a group of patients with traditional amputations, the first patient to have AMIs implemented in his residual limb has improved control over a robotic prosthesis. This AMI patient reports experiencing natural sensations of ankle-foot positions and movements even when blindfolded. He also displays emergent natural reflexive behaviors while walking up and down stairs, as well as improved performance on tasks requiring control over how hard he is pushing on a foot pedal. In addition to these functional improvements, the AMI patient also moves and behaves as though the bionic limb is part of him, which provides evidence of neurological embodiment. Although this is a single-patient case study, the results highlight the potential of the AMI to help persons with amputation connect more completely with their prosthetic limbs.
We have also observed a lack of atrophy and aberrant phantom sensation in our cohort of AMI patients. Although these results are quite preliminary, they indicate that the AMI may provide benefits over traditional amputation even in the absence of an advanced robotic prosthesis.
The AMI was invented at the MIT Media Lab [see relevant patents], by a research team comprised of surgeons, scientists, and engineers. Given that the AMI involves both designed tissues and synthetics, the development team spanned many scientific and clinical disciplines.
To date, AMIs have been surgically implemented in nine patients.
The Ewing Amputation is the name given to the surgical procedure in which AMIs are constructed from native muscle tissues at the time of primary below-knee amputation. It was named after Jim Ewing, the first patient to undergo the operation. All nine patients to receive AMI amputations thus far have undergone the Ewing Amputation.
Thus far, the AMI has been implemented only at the below-knee amputation level in persons requiring an elective amputation. However, it is important to note that the benefits of the AMI are not restricted to this patient population. Research is already underway to explore construction of AMIs at the above-knee amputation level, as well as in the persons with amputated arms and hands. The AMI is not just for legs!
It is important to note that implementation of the AMI may not be appropriate in patients requiring amputation due to advanced peripheral vascular disease. Patients in this population typically exhibit neuropathy and microvascular compromise, which may negate the benefits of the AMI and inhibit proper wound healing. Nevertheless, even if this population were excluded entirely, a majority of the remaining estimated 46% of patients indicated for amputation would be eligible for an AMI procedure.
In short, yes. A recent pre-clinical study demonstrated that it is possible to leverage regenerative capabilities of nerve and muscle tissue to build AMIs even in cases where distal tissues are no longer available, such as traumatic amputations or revisions to existing amputations. This approach has not yet been implemented in human patients; we expect to run those trials in the near future.
If you have a medical indication for a lower-extremity amputation, you might be eligible for on-going research trials. To learn more, visit our page on clinicaltrials.gov or call 617-983-4555. To reach Professor Herr and the AMI research team, contact ami-team@media.mit.edu.
The AMI has been described in several publications, in both pre-clinical and early clinical studies. To learn more, see the Publications page.