Algorithms can be used to design cheap, perfect-fitting limbs, orthodontics and other accessories.
In 1981, Ian Gregson popped into the prosthetics shop in a Vancouver hospital to buy a new leg. The shopkeeper took various measurements, then made a plaster cast of the amputee’s residual limb to craft an attachment.
“The leg was just trash. It was 2 inches short, with a suction socket, and no padding,” Gregson says. “It was very painful. I tried and tried, but it didn’t work.”
Prosthetic limbs, particularly the sockets that hold them on, must be carefully fitted to an amputee. Now, a stunt man in the film industry and a two-time competitor at the Paralympic Games, for shot put and discus throw, Gregson describes his current socket-maker as an artist, a master who passes his skills on to a few select protégés.
Gregson’s favored quad socket has a rounded rectangular opening with a silicone liner. After a test socket confirms the fit, the real thing is cast, followed by sandpaper and a heat gun for adjustments. Gregson insists the personal touch is what really makes the difference, and there will always be something that needs tucking in or letting out. Even the way someone stands during a measurement—if his hips aren’t completely level—can throw off the fit.
» Get the best federal technology news and ideas delivered right to your inbox. Sign up here.
Expert prosthetics makers are rare animals, however, and cannot be mass-produced, unlike technology. While an exact fit via traditional methods calls for almost virtuosic expertise, the emergence of 3-D printing may allow for much more precise appliances—ones that could be made right at home. Bespoke 3-D printouts could even disrupt orthodontics and other common correctives.
The human leg has a more complex structure than any external cast can discern, and it changes with every motion and flex. David Sengeh of the biomechatronics group at MIT Media Lab, is working to convert the iterative process of prosthetics-building into a one-time procedure by focusing on amputees’ internal anatomy.
He begins by mapping a remaining limb using Magnetic Resonance Imaging, which reveals the shape and consistency of its underlying tissue. More accurate than traditional caliper- or laser-scanning techniques, the MRI distinguishes among bone, muscle and articular cartilage and avoids the alignment problems that come from an external scan.
Using the tibia as a reference point, computer-aided design is combined with an engineering technique called finite element analysis to predict which internal pressure points on the residual limb could most comfortably bear the load of the body’s weight. An algorithm designs a socket flexible in some areas and rigid in others, using a variety of materials.
“We’ve been able to make the [world’s] first socket entirely from quantitative methods,” Sengeh says. “No human hands were involved in defining the shape, including the cut lines and material properties of the socket.”
Throughout the 1990s, Sengeh’s home country of Sierra Leone suffered a civil war, where around 8,000 people had one or more limbs violently amputated. Many of those amputees had access to prosthetics but abandoned them because of the discomfort of a bad fit.
In contrast, Gregson was run over by a train in a part of England that didn’t often see that type of traumatic injury. The surgeon had done his best, but Gregson’s residual limb was left boney, without the fleshy parts that can make a socket more comfortable
Sengeh’s idea is that a good-enough algorithm can eventually reproduce the expertise of a human prosthetist. Anywhere in the world, an amputee can send Sengeh a minimal set of data, and he can mail back a comfortable prosthetic socket. Alternatively, a 3-D printer in the amputee’s town could produce the socket on the same day the measurements were taken.
The next step would be to scale up the project, but until then, amputees will continue to fit their prosthetics to purpose.
“Even I, with access to first-class health care, use duct tape on it all the time. I’d be in the shop every week if I went in to fix every little squeak,” Gregson says. “The biggest issue comes if I gain or loose weight. The,n the socket becomes too tight or too loose.”
“People always have their own solutions—experimenting with themselves and hacking their own prosthetics,” agrees Sengeh.
Perhaps the greatest promise lies in combining those two approaches: the high-tech equipment and the human ingenuity that people bring to their own problems. For example, Amos Dudley, a college student studying digital design, recently fixed his own protruding and overlapping incisors at about 1 percent of the cost of orthodontic treatment.
As an undergraduate, Dudley had very little spare cash but one significant asset: access to New Jersey Institute of Technology’s fabrication labs, which included a high-resolution 3-D printer, computer-assisted design software and a vacuum forming machine.
After taking a cast of his own teeth and plotting a route for each one, Dudley created 12 clear aligners to bring his teeth incrementally closer to a perfect smile. He bought retainer plastic on eBay, and his total outlay was only $60.
“They’re much more comfortable than braces, and fit my teeth quite well,” Dudley writes on his personal blog. “I was pleased to find, when I put the first one on, that it only seemed to put any noticeable pressure on the teeth that I planned to move.”
After 16 weeks, the misalignment was fixed, but Dudley continues to wear the final aligner at night to guard his teeth against the damage of grinding. It provides the perfect fit as a whitening tray. (Dentists caution the aligners can only solve for rotation and tilt problems in simple cases.)
Assuming 3-D printers follow the pricing pattern of new technologies—cheaper with time—the cost difference between professional and amateur manufacturing will only grow in the coming years. In the future, distance from a specialist or lack of resources may be no barrier to comfort.