by E. Paul Zehr
Figure 2.7. The circle represents the body center of mass suspended over the legs, which work as springs while walking (A); the “leg spring” comes from controlling muscles that cross the hip, knee, and ankle (B); exoskeleton worn over the leg used to change the mechanical responses of the ankle joint (C). Courtesy Daniel P. Ferris.
Figure 2.8. Parade performer in Disneyland, California, using a lower limb exoskeleton to amplify the springlike activity that the legs produce normally.
The bottom line of all the scientific analysis is that the carbon fiber “blades” could significantly enhance performance. This is largely because the blades allow for a far more efficient running pattern. The blades are actually much lighter than the lower legs they replace, which means that the legs can be moved about 15% faster than the highest performance of sprinters with intact legs—including 2008 double gold medal winner Usain Bolt of Jamaica. Also, the same overground running speeds could be obtained using the blades while applying about 20% less force into the ground. Overall the “springiness” of the blades meant that only about one-half of the muscle force needed for sprinting at the same speeds with intact limbs was needed with the prosthetic legs.
If Oscar were a marathon runner, there would be different issues. On New Year’s Day in 2010 American ultradistance runner Amy Palmiero-Winters won the “run to the future.” She covered 130.4 miles in 24 hours, making her the first person with a prosthetic lower leg to qualify for the U.S. track team. Her situation is different from Pistorius, because the mechanical benefits Amy might get from a single prosthetic leg don’t really help with long-distance running.
If you keep your eyes open, you can see small-scale applications of assisting human movement with technology in many different places. I took the picture of a performer during a family trip to Disneyland in 2009 (figure 2.8). The basic ideas we have been discussing are clearly shown in how he used spring boot pogo sticks to amplify movement. The main point of this as it relates to Iron Man is that even simple devices can augment and improve function in people with amputation—and those like the Disneyland performer who are just trying to have fun.
What is important to consider is that really efficient machine-based locomotion should probably mimic what we do when we just walk around. So, an Iron Man armored suit should do the same. After all, as Tony Stark said while testifying at the “Weaponized Suit Defense Program Hearings” (shown in the Iron Man 2 movie), his device is really just “a high-tech prosthetic.” Now let’s look at some other prosthetics that link directly to the nervous system.
CHAPTER THREE Accessing the Brain of the Armored Avenger
CAN WE CONNECT THE CRANIUM TO A COMPUTER?
Undergoing the Extremis Procedure remade my body from the inside out. Long story short, my body was turned into a kind of computer designed to interface with the Iron Man. There was no longer a division between me and the suit. My brain … evolved, I guess. Into a kind of hard drive.
—Tony Stark, “World’s Most Wanted, Part 2: Godspeed” (Invincible Iron Man #9, 2009)
The original version of the Iron Man armor was designed to preserve my damaged heart. The obvious next step was to extend the suit’s preservative capabilities to an even more critical organ…. Not so much my brain per se, as my cognitive neurofunctions and basal personality structure.
—Tony Stark speaking out loud about new changes to the armor, from “Hypervelocity #2” (Iron Man, 2007)
Coffee pot on!” Imagine if you could, upon awakening, simply have that thought and your coffee maker would go on in your kitchen. Or your kettle, if you prefer tea. Imagine if your thoughts could be transformed into the actions of machines. In the last chapter, we looked at muscles and how they make us move after receiving electrical commands from the nervous system. In this chapter, we look at how we can tap into the command signal from the nervous system to not just make muscles contract but to trigger powerful motors to move robotic suits of armor. Another focus of this chapter is on what happens when that chain of command is broken, which relates to that imaginary ability to turn on your coffee maker with thoughts alone. We will discuss how that works by giving some examples of prosthetic arms and legs being powered with the mind.
These real-life bionic men and women bring us to the fascinating field of neuroprosthetics. This term refers to devices that are used by or implanted into a person to improve sensory, and, in some cases, cognitive abilities, including retinal and cochlear implants to augment sight and hearing, spinal implants to relieve pain among other things, and implants to assist with bladder control. Something that Tony Stark would be able to use would be the implants that are being developed to control movement of an object by simply thinking about it. These inventions are in their infancy but are expected to help paraplegics and quadriplegics and others with severely limited movement. And who knows what else they might soon be able to do?
Some of the possibilities of the future are being revealed by pioneers such as Kevin Warwick, a professor of Cybernetics at University of Reading in the United Kingdom. Warwick is known for many things including “Project Cyborg”—his attempt to have implants placed into his body that can be used to control other devices. One of his early efforts was to have a computer chip implanted that could be detected by sensors outside his body to then turn on lights or other appliances. Later, he had an electrode array implanted in his arm. This array took information from a nerve in his forearm, which was used to control a robot arm “directly” and eventually over Internet relays. Warwick summarizes his approach in his book I, Cyborg. He has had some exciting successes with this approach, but the problems that have arisen in doing more complex tasks for longer periods highlights how difficult interfacing humans and machines actually is.
Remember from chapter 2 that when you want to do something, you must activate your muscles. When you integrate your body with a robotic machine, you must skip the muscles and go directly from the output in the spinal cord straight on to the machine. In a way, we are doing some wire-tapping in the nervous system. The first wire tap we want to set is from the spinal nerves. Later, we will also talk more about even tapping into commands from the brain itself using a special kind of neuroprosthetic—a brain-machine interface.
The general principles for neuroprosthetics, as well as other types of prosthetics, are similar for both amplifying human performance and for replacing it. Many advances in prosthetics have improved the “usability” and the look of the devices. As we saw last chapter, however, the limbs themselves don’t actually function in the same way as in an intact situation. An idea that arose early on in the field of neuroprosthetics was controlling a motorized limb using commands from the nervous system. In other words, tapping into the commands that would normally activate the muscles themselves. Instead of needing to figure out exactly what the complex sets of commands should be for a given movement, a simple and elegant approach is to instead just use the input itself.
Most neuroprosthetics detect signals from the person’s nervous system and relay these signals to an electrical controller inside the device. The biological signals could be electrical activity of muscle detected from electrodes on the skin or implanted in muscle, nerve signals detected by implanted electrodes, or even electrode arrays in the nervous system that have the nerve cells growing through them. In this way the controller of the neuroprosthetic is literally connected to the neuromuscular system and to the device. The commands from the person can then be detected and relayed to the device to make it do whatever it is supposed to do to replace the lost function.
Monitoring Muscle and Nerve to Make Motors Move
If damage to the nervous system, such as an injury to the spinal cord, occurs, electrical stimulation can make the muscles contract even when the nervous system itself cannot provide the command for the contraction. Since the point of the stimulation is to help with functional movement, a term that arose is “functional electrical stimulation,” or FES. FES is now more broadly used to refer to using electr
ical currents to produce or suppress activity in the nervous system. When similar stimulation concepts are used to more generally alter activity within the nervous system, it is often called “therapeutic electrical stimulation.”
An example of a really useful FES device is the “WalkAide.” I bet you would never guess it helps with walking. The WalkAide is a small battery-powered device used to stimulate a nerve that activates the muscles that help flex your ankle to bring the top of your foot up when you walk. If your nervous system is working well, you probably pay this no attention at all, but the clearance of your foot over the ground when you walk is a hugely important issue during walking.
Because it is energy inefficient to pick the foot way up off the ground while walking, your nervous system activates your leg muscles so that the bottom of your foot just clears the ground by less than an inch. This is all fine and good until something like a spinal cord injury or a stroke occurs. These disorders lead to weakness of the muscles, particularly those that flex the ankle, and suddenly walking is much more difficult. You may have experienced a brief example of the outcome of this phenomenon, called “foot drop” or “drop foot.” Sometimes a piece of sidewalk will be broken and jutting up higher than the clearance of your foot while walking. You don’t see it as a major object visually but then you scuff your foot and may trip. Well, after a stroke this is common even when the place a person walks on is level and smooth. Two simple things can be done: one is to swing the leg out and around when walking (so the affected leg arcs to the outside and forward—called “hip hiking”). The other is to wear an external brace of metal, plastic, or graphite that holds the foot at about a right angle so it cannot scrape against the ground. Neither is really a great approach. So where does FES fit in?
With the WalkAide (figure 3.1), based on the acceleration and the angle of the leg, a sensor detects the correct time during walking when flexing the ankle should occur but doesn’t because the nervous commands are lacking due to injury. So, the WalkAide applies electrical stimulation to the nerve, activating the muscles that flex the ankle and allowing the person to pick up the foot. If you reach down the outside of your knee, you will come to a little bump on the outside of your lower leg just below the knee. This little bump is actually the “head” of the fibula—the leg bones we talked about earlier with Oscar Pistorius. Right near the head of the fibula is the common peroneal nerve that innervates the major flexor of the ankle, the tibialis anterior muscle (see figure 2.1). The WalkAide appropriately stimulates (and just as importantly) stops stimulating this nerve in order to get the foot moving better during walking. It is small, easy to use, and requires 1.5 volt AA batteries. The stimulator unit is shown in panel A and a drawing of how it would help move the ankle by stimulating the ankle flexor muscles is shown in panel B. The bulk of the initial WalkAide technology was developed in the laboratory of Richard Stein at the University of Alberta in Edmonton, Canada, during the late 1980s and early 1990s. I was doing my doctoral training in neuroscience there in the 1990s and got to see some of the early prototypes from this work up close. It was captivating then and still is now. But back then it was an idea in development. Now it is an actual product that people can get and use to help improve their walking.
Figure 3.1. WalkAide neuroprosthetic stimulator, which corrects a condition known as drop foot. The stimulator (A) activates the common peroneal nerve, which causes activity in the muscles that lift the foot (B). Images courtesy Innovative Neurotronics.
The WalkAide is an example, then, of a neuroprosthetic. It represents a medical device that helps improve (or replace in some cases) bodily function that has been lost due to accident or disease. Other kinds of neuroprosthetics include stimulators for bladder and bowel control, deep brain stimulation (we will come back to this later on in this book—so stay tuned), and cochlear prosthetics. Usually neuroprosthetics require insertion and implantation of electrodes into the body near the nerve or muscles that are targeted. But, the WalkAide is an example of a neuroprosthetic that doesn’t need any implantation.
Cochlear prosthetics to improve hearing are the most commonly applied and utilized neuroprosthetics. They represent an instructive example of how continuing evolution in the fields of biomedical engineering and neuroscience from the 1950s to now have dramatically improved neuroprosthetic devices. Originally cochlear implants were very large and had external pieces fixed to the body that were wired to parts implanted into the inner ear. Now they are small directly implanted devices. This really nicely parallels the concept of the cardiac pacemaker and defibrillators. We will discuss this in chapter 7 when I put a new twist on Iron Man’s origin story!
Another interesting example of this kind of FES is a neuroprosthetic for improving hand function developed by Arthur Prochazka (coincidentally also at the University of Alberta). This “bionic glove” helps people who have problems moving the wrist and hand due to a stroke or spinal cord injury. As long as the person has some ability to move the wrist, the glove can help stimulate the muscles in the forearm that control grip. Sensors detect wrist angle and then trigger small stimulators to electrically activate the flexor muscles of the forearm. Imagine picking up a bottle of water. As you reach out and pick it up you first open your hand, make contact with the bottle, then close your hand, and lift it up. If you had partial paralysis of your arm and hand this would be difficult, if not impossible, to do. With the bionic glove, the user reaches out and gets the hand around the bottle. But because she cannot contract the flexor muscles, she extends the wrist a bit more. This signal triggers the glove to flex the fingers and the grip is made. Then the bottle can be picked up and used. In this simple device, three different muscle groups are activated with electrical stimulation.
So far our examples have had to do with neuroprosthetics that detect signals related to a residual movement that someone can make (like a bit of a walking movement, a bit of a wrist movement). Then the devices use that signal to trigger a stimulator to activate muscles that cannot get the normal activation from the nervous system. It is important to understand how these types of devices function to get closer to appreciating how the Iron Man suit could actually work. For starters, the suit would have sensors detecting nervous system commands from nerve or muscle as well as commands from residual movement. The suit then would amplify the normal movement. However, it wouldn’t do so by stimulating the muscles like in FES. Instead the trigger signals would drive the motors controlling the joints in the Iron Man robotic suit. Above we were talking about restoring function in a damaged nervous system with FES and neuroprosthetics. In that way the neuroprosthetic helps “bridge” the problems in the nervous system to restore some movement ability. This is what Tony will need to operate the NTU-150 and already exists in the form of the Cyberdyne Hybrid Assistive Limb (HAL) wearable robot suit.
HAL is a kind of robot suit that is worn in order to improve physical capability. As we already learned, when someone tries to make a movement, weak electrical signals travel in the nerves and occur in the muscle during a contraction. These weak signals can be detected, measured, and amplified with electrode sensors placed on the skin over the muscles being used. The HAL suit uses this control signal to trigger the control of motors acting at joints on the suit. As a result, the suit is controlled directly based on the commands coming from the person wearing it. So, controllers for the elbow joint motors are triggered from nervous system commands going to the muscles that normally flex and extend the elbow. Cyberdyne Inc. likes to call this a “voluntary control system.” This type of system relies on the users’ intended movements to then amplify those movements by making the robot suit do the appropriate action. An additional layer of control is added using a “robotic autonomous control system,” which is a kind of predictive system that works along with the voluntary triggering. All together, HAL applies a hybrid of the two control modes that provide an almost human-like movement. We will pay more visits to HAL later on in the book.
This basic concept o
f hybrid control has also been used by a company called Touch Bionics in their development of a fantastic neuroprosthetic hand. Think back to our medieval “Iron Hand” prosthetic shown in figure 2.4. Touch Bionics has created a sophisticated robot hand prosthetic that is driven by the normal muscle activation signals for the fingers. It can also be controlled by touch signals taken from pressure sensors. The Touch Bionics 5 finger i-LIMB hand uses inputs that come from the normal muscle signals to open and close the lifelike plastic fingers in the prosthetic. So, it uses the signals that come from muscles in the stump or remaining part of the person’s arm. The i-LIMB hand then can open and close to grasp objects in a way similar to a biological hand (panel A of figure 3.2).
Figure 3.2. Touch Bionics 5 finger i-LIMB hand, which uses inputs that come from muscle signals to open and close the lifelike plastic fingers in the prosthetic. The i-LIMB makes a pinch grip (A) and individual “Pro-Digits” can be used for people with partial amputations (B and C). Courtesy Touch EMAS Ltd.
ProDigits is an application of this device for people who are missing one or more fingers due to accident or from birth. This device has individually powered and controlled motors for each finger and can be set up to take over for just the fingers needed by the user. This means that a lot more than just an open and closed grip can occur and more dexterous activities can be done, such as pointing with the index finger and typing on a keyboard. These seem pretty simple tasks—and they are if you have an intact hand. But they are not if you don’t. An example of replacing one finger is shown in panel B of figure 3.2 and replacing function for four fingers is shown in panel C. The i-LIMB hand and the ProDigits can be covered in a flexible skin product making it look just like a real biological hand. Or they can be left uncovered. Tony Stark would go for the covered option if he needed one, I think.
