Replicating the Ideal Gait Guides Design of New Prosthetic Foot

MIT researchers developed a quantitative framework that enables them to replicate biomechanical performance of prosthetic feet, a new approach that could lead to inexpensive mass-production of the prosthesis.

by Poornima Apte
July 30, 2018

A team of engineers at the Massachusetts Institute of Technology has developed a passive prosthetic foot that can be customized to individual needs while still keeping costs low.

You can find many prosthetic models on the market, but too often they’re powered by technologies that make them expensive and out of reach for many people, especially those in the developing world.

Jaipur Foot, a prosthetic limb manufacturer in Jaipur, India, has developed a solution. In 2012, the manufacturer approached Amos Winter, assistant professor at MIT’s Department of Mechanical Engineering and director of the Global Engineering and Research Lab, to spearhead research into a low-cost alternative to their existing prosthetic feet. Jaipur Foot was already outfitting amputees with prosthetics, but these were heavy and handmade, which made them difficult to reproduce on larger scales.

For You: A More Natural Approach to Passive Ankle Prostheses

Most designers create prosthetic feet based on a lifelike model. Winter and his team, though, decided to turn that logic on its head and work backward. Instead of focusing on reproducing exact physiology, what if they concentrate on reproducing an able-bodied gait? This would involve designing a range of cheaply produced “feet” and testing which one comes closest to delivering the desired gait.

For the best solution, Winter and his team had to figure out how varying the mechanics of the foot - such as stiffness and thickness - would affect how the lower leg moves. They started with existing research that monitored steps taken by able-bodied people of different heights and weights. The ground reaction forces generated by each step were especially useful to Winter’s team.

Using this data, and incorporating what the ground reaction forces should look like for people of certain heights and weights, they tailored the foot accordingly. The mathematical model they developed describes the stiffness, possible motion, and shape of the foot. By plugging in the reaction forces data, they predicted the range of motion that different feet would deliver.

“The size of the foot is dictated by the user; we want the foot length to match their physiological foot length,” Winter says. “The stiffness (and thickness) of the foot is dictated by our design framework, where we can tune the compliance of the foot so it bends enough during walking to facilitate near able-bodied motion and loads.”

Next came the shape. To arrive at the best foot shape, the team parsed the options through a “genetic algorithm” and figured out which one was the “fittest.” A large number of feet with different curve shapes was plugged into simulation. The ones with a high error in delivering the desired trajectory were discarded. The final result, made of machined nylon, looks somewhat like a toboggan.

The final results, recently published in ASME’s Journal of Mechanical Design, are “pretty darn close” to able-bodied gait but not completely there. Since adoption of these toboggan-shaped feet will require them to look more life-like, Winter and team are working with Vibram, an Italian company that manufactures outsoles, to develop life-like covers for the prosthetics. The covers are also expected to deliver better traction, something the feet alone don’t have much of.

The team is working on on-the-ground implementation strategies, including having a lookup table where a technician could narrow down which model would work best for a person’s weight and height and take it from there.

“Particularly in limb fitment camps, where a patient may interact with a technician for only a matter of minutes to have a foot fitted, there is a lot of value in a lookup table where the technician can choose a foot based on the patient size and bodyweight,” Winter says. “But we also have to account for user-centered preferences. Some people may prefer a stiffer or more compliant foot, so there might be some iteration in the process. The lookup table could provide an accurate first guess, and then the prosthetist could use their experience and patient feedback to dial in the correct stiffness.”

The level of customization the team’s feet deliver is particularly important to female soldiers and veterans. “There is a congressional mandate to improve the customizability of prosthetic feet for this population, who too often have to wear men’s feet that are too big and stiff,” Winter says.

Although the feet are customized, they can still be inexpensive and mass-produced, since the level of customization, including the use of different materials, varies depending on the final cost structure.

“For developing world markets, we will likely have different gradations of feet in size and stiffness. The prosthetist would find the foot closest to ideal for a patient,” Winter says. “In wealthier markets, we are confident we can still make pretty low-cost, specifically customized feet if they are made of plastic, like our current foot is made of nylon. The feet can be custom machined for an individual. For higher segments of the market we could possibly also offer custom carbon fiber feet, which would be lighter but obviously more expensive.”

Most prosthetic feet produced today are passive and “designed using qualitative or empirically-derived metrics, and are not quantitatively designed for a desired biomechanical performance,” Winter says. His are.

Poornima Apte is an independent technology writer.

Read More from How to Hack a 3D Printer to Print a Human Heart

Device Could Allow Chemo Patients to Test Blood Cell Levels at Home

Fiberscope Lets Scientists Look Deeper Into the Brain