Skip to content

Bioinspired Artificial Skin Sensor Warns Prosthesis Users of Dangerous Pressure

Healthcare professional examining a man with a prosthetic leg and highlighting knee pain with digital overlay in clinic.

A prosthetic limb may enable people to stand, walk or run again, yet it cannot always alert them if its socket is exerting excessive pressure on the skin of the residual limb.

This is important because too much pressure, or pressure distributed unevenly at the point where a prosthesis connects with the body, may lead to discomfort, inflammation and ulcers. It may also alter a person's balance and mobility.

Researchers in China have developed a bioinspired sensor designed to pinpoint pressure and recognise patterns that may indicate potential damage.

Reported in a study in Cyborg and Bionic Systems, the device was trialled on a robotic hand and in a small feasibility study with people who had lower-limb amputations.

Artificial skin for prosthetic sockets

Rather than being a full prosthetic limb, the prototype is a sensing layer intended to fit within the socket, between the prosthesis and residual limb. Functioning as artificial skin, it tracks where the socket applies pressure, its force and its duration.

Short-lived, 'harmless' pressure is interpreted as normal touch.

Pressure that persists, recurs or increases over multiple locations, however, may produce an early alert that tissue is under threat.

The sensor operates through two complementary pathways. Its haptic pathway quickly communicates the position and strength of pressure, while its pain-inspired pathway integrates signals across time and between sensing locations.

It incorporates pressure-sensitive material and synapse-like transistors – electronic parts that can alter their response based on signals they have received before.

That arrangement gives the system a basic type of memory. After an injury-like stimulus, it becomes more sensitive in its warning response, meaning lower pressure may later generate a stronger signal.

"The device neither feels pain nor creates pain in the user," senior author Huaping Wu, a mechanical engineer at Zhejiang University of Technology, told ScienceAlert.

"Instead, it reproduces selected information-processing features associated with nociception, including stimulus thresholds, temporal and spatial summation, memory, and sensitization."

Pain in living organisms is substantially more complicated, encompassing nerves, the spinal cord, brain, emotions and conscious awareness. This device delivers a practical warning rather than pain.

Testing pressure warnings on a robotic hand

The research team initially attached an array of four sensors to a robotic hand. Pressure of 10 kilopascals lasting 0.21 seconds was classified as harmless by the system. Yet when that same pressure continued for 0.49 seconds, the signal built up, exceeded the warning threshold and caused the hand to pull away.

Following this injury-like occurrence, identical pressure triggered withdrawal within 0.06 seconds. Although this replicated the increased sensitivity seen in the body after injury, it did not indicate that the robot consciously remembered anything or experienced pain. Its quicker reaction resulted from an engineered shift in the warning threshold.

Prosthetic limb socket study

The team next installed sensors inside the below-the-knee prosthetic sockets of three participants, measuring pressure while they sat, walked, climbed stairs, jumped and ran.

Four sensing points were positioned in areas prone to pressure, chosen according to anatomy and the participants' accounts of day-to-day discomfort.

While running, the system identified excessive lifting of the hip used to help the prosthetic foot avoid the ground. This created uneven loading and raised pressure at the front of the residual limb. Feedback from the device enabled participants to alter their gait and lessen the excessive hip movement.

To assess the device's warnings against human evaluations, researchers used the participants' numerical discomfort scores alongside observers' ratings of facial expressions. By combining spatial and temporal signals, the system correctly detected 86.4 percent of short discomfort events and 90.9 percent of prolonged events.

These results reflect agreement during the tests rather than satisfaction among users. As the paper does not state the overall duration for which participants wore the system or provide a satisfaction survey, it cannot yet establish whether people would regard it as comfortable or practical for everyday use.

Other prostheses have returned feedback in other ways.

A bionic hand linked to a user's bones, muscles and nerves supplied limited touch, whereas a sensory prosthetic leg converted pressure under an artificial foot into signals delivered to nerve endings.

By contrast, this device observes the concealed interface between the residual limb and socket, where prolonged or uneven loading may harm tissue.

Its alerts could encourage users to reposition themselves, modify their gait, take a rest or get the socket inspected. Logged pressure patterns might also assist clinicians in tailoring socket fittings and gait training.

Limits of the prototype

The findings are still preliminary, and the wired prototype needs to be miniaturised into a compact, low-power wireless system.

"Medical-grade protection, secure attachment, and stability under repeated loading, sweat, temperature changes, and socket deformation must also be established," Wu told ScienceAlert.

Bigger, longer-term studies covering different amputation types and socket designs will be required to establish whether the system enhances comfort, tissue protection and rehabilitation.

Should it prove dependable, this artificial skin may alert prosthesis users to hazardous pressure before discomfort develops into injury.

The study was published in Cyborg and Bionic Systems.

This article was fact-checked by Rachel Garner and edited by Rebecca Dyer. While we pride ourselves on our process, we are only human. If you spot a mistake, please let us know.

Comments

No comments yet. Be the first to comment!

Leave a Comment