BMI-Avatar-Robot-Training Wakens SCI Recovery

Posted by Sam Maddox in Research News on August 15, 2016

In an experiment in Brazil, eight paraplegics, seven considered to have complete spinal cord injuries, used a brain-machine interface (BMI) technique to use their thoughts to move their legs in two kinds of robotic devices. Not news, we’ve heard of this sort of thing for years.

What is news is that by using a non-invasive BMI skullcap, along with a cool virtual reality training method, each of the participants in the year-long program recovered partial neurologic function, both sensory and motor.

The work comes from physician scientist Miguel Nicolelis, who is from Brazil, and who has a position as co-director of the Duke University Center for Neuroengineering. Maybe you remember him; we’ve covered his work several times, in 2013, in 2014, and of course including the stunt he pulled off two years ago – one of his paralyzed subjects, in an exoskeleton, kicked a ball to start the World Cup in Sao Paulo. It wasn’t much of a kick, to be honest, but it drew attention.

The story of the eight recovering paras was published last week. This approach, says the paper, resulted in some kind of spinal cord plasticity – the remodeling of signals of brain and spinal cord nerves. This is from the paper, in the journal Scientific Reports, an open access journal from the publishers of Nature (note: it is available online in its entirety):

Patients ... regained voluntary motor control in key muscles below the SCI level, as measured by EMGs, resulting in marked improvement in their walking index. As a result, 50 percent of these patients were upgraded to an incomplete paraplegia classification. Neurological recovery was paralleled by the reemergence of lower limb motor imagery at cortical level. We hypothesize that this unprecedented neurological recovery results from both cortical and spinal cord plasticity triggered by long-term BMI usage.

"For the first time in many years they were able to voluntarily control their muscles," said Nicolelis. "They could move their legs or contract muscles under voluntary control." Nicolelis told reporters that some of the group of eight continue to improve. He said two women, paralyzed for more than a decade, "can generate leg movements, move their legs out and in and flex their knees."

Here’s Nicolelis, from a Duke press release, which I recommend for further background:

What we're showing in this paper is that patients who used a brain-machine interface for a long period of time experienced improvements in motor behavior, tactile sensations and visceral functions below the level of the spinal cord injury. Until now, nobody has seen recovery of these functions in a patient so many years after being diagnosed with complete paralysis.

So what’s going on here? Nicolelis suggests that BMI is not just an assistive device to move a prosthesis, as numerous experiments have shown to be true, but that in combination with other therapies, BMI may indeed become a new therapy in its own right.

How so? He’s not sure exactly how it works, but Nicolelis thinks repeated BMI activity could engage dormant spinal cord nerve networks, the so-called central pattern generators (CPG). Awakening these circuits is, of course, the primary hypothesis that underpins spinal cord stimulation, and thus the epidural stimulation success in Louisville with recovery in a handful of motor-complete paraplegics; firing up the CPG is also the theory behind the upcoming Big Idea trial.

But something else besides the CPG may be at work here: Perhaps the virtual reality part has encouraged some sort of cortical reshaping. In the Nicolelis study, each participant donned a fitted-cap with 11 electrodes to record EEG activity. They used the brain signals to learn how to move the legs on a virtual reality avatar; those same thought patterns were basically hacked, fed into a computer, and used to power robotic legs on either a Lokomat (robotic treadmill walker) or an exoskeleton. So far, that’s the usual BMI story.

But the avatar training also included a tactile feedback system, transmitted to the user’s arm. According to Niclolelis, when subjects were asked to imagine taking steps in their virtual world, his team was not able to measure signals in the brain region normally associated with leg control. “If you said, use your hands, there was modulation of brain activity," Nicolelis said. "But the brain had almost completely erased the representation of their lower limbs."

But after several months of training, he said, brain activity for leg movement reappeared in the part of the brain you’d expect to find it. "Basically, the training reinserted the representation of lower limbs into the patients' brains," Nicolelis said.

Here’s more on the feedback piece, from Duke:

During most of their training, the participants also wore a sleeve equipped with touch-technology called haptic feedback to enrich the experience and train their brains, Nicolelis said. Haptics use varied vibrations to offer tactile feedback, much like the buzzing jolts or kickbacks gamers feel through a handheld controller. Each sensation is unique. So when the avatar walked on sand, the patient felt a different pressure wave on the forearm than when they walked on grass or asphalt, Nicolelis said.

"The tactile feedback is synchronized and the patient's brain creates a feeling that they are walking by themselves, not with the assistance of devices," Nicolelis said. "It induces an illusion that they are feeling and moving their legs. Our theory is that by doing this, we induced plasticity not only at the cortical level, but also at the spinal cord."

Clinically relevant? No, but maybe it could be if they prove that repetitive BMI has a beneficial biological effect, or that avatar tactile feedback actually rewires the brain. Also, the benefits of long-term training, which we know have an effect on recovery, need to be explained in future studies – maybe the training aids plasticity and not BMI at all. So, it’s always good to report recovery in chronic SCI but let’s not overstate the significance of BMI plasticity until the researchers pin down which of their interventions is most relevant to recovery, and why.

Nicolelis wants to repeat the study in sub-acute patients, those with injuries just a few months old. From the paper:

... functional cortical plasticity likely led to the reactivation of upper motor cortical neurons that normally project to the spinal cord, via the corticospinal tract.

Particularly, in the case of locomotion, the use of robotic gait training may have triggered the engagement of central pattern generators (CP), both at the supraspinal and spinal levels, and contributed to the generation of tactile and proprioceptive feedback from the patients’ own legs. These two components may have also induced functional plasticity at the spinal level and contributed to the type of sensory recovery observed below the level of the SCI, mediated mainly by the spinothalamic tract, and the somatosensory cortical plasticity described above.

Similarly, we hypothesize that long-term training observing a human avatar mimicking the position and orientation of the patients’ body could have induced a positive effect on our patients’ sensory acuity.

Therefore, the present findings raise the relevance of BMI-based paradigms, regarding their impact on SCI patient rehabilitation. In this context, it would be very interesting to repeat the present study using a population of patients who suffered a SCI just a few months prior to the initiation of BMI training. We intend to pursue this line of inquiry next. Based on our findings, we anticipate that this population may exhibit even better levels of partial neurological recovery through the employment of our BMI protocol.