What “bionic” means here
“Bionic” is partly a marketing word. In prosthetics it is usually applied to arms or legs that have motors and electronic control, as opposed to purely mechanical ones. The most common form of control for arms is myoelectric: the device reads the small electrical signals produced when the user contracts muscles in the remaining limb, and uses them to drive small electric motors. Before this, the usual approach was a cable and harness driven by shoulder movement.
The idea behind myoelectric control
When a muscle contracts, it generates a tiny electrical signal that can be detected by electrodes resting on the skin. Researchers realized in the early twentieth century that these signals could be amplified and used as a command. Even after an amputation, the muscles in the remaining part of the limb still contract when the person intends to move the missing hand. A simple system can compare the signals from a pair of opposing muscles, such as those that flex and extend the wrist: one signal means open, the other means close.
Turning this principle into a working device needed small amplifiers, motors and power sources. Those only became practical after the transistor, which put the first workable systems in the middle of the twentieth century.
Early experiments
Accounts differ on who built the first myoelectric arm. A German physicist, Reinhold Reiter, is often credited with a prototype in Munich in about 1948, but it did not reach regular use. In the Soviet Union, a team led by Alexander Kobrinski developed a myoelectric hand around the late 1950s, which was publicized abroad, and the term “Russian hand” was used for it. Researchers in Britain, Canada, Sweden and the United States followed, and several hospitals tested their own versions in the 1960s. The early devices were heavy, used large batteries, and often failed in everyday conditions.
In the United States, a project at the Massachusetts Institute of Technology and a Boston hospital in the late 1960s produced the Boston Arm, a myoelectric elbow that moved smoothly in proportion to the signal. Later, in the 1970s and 1980s, a group at the University of Utah developed a powered elbow that became known as the Utah Arm. These designs helped establish the powered elbow as a normal option for people who had lost an arm above the elbow.
Control improves
The earliest myoelectric hands had one motion, open and close, and a single pair of signals. Over the following decades, the electronics improved. Instead of a fixed on-off switch, controllers began to respond proportionally, so that a stronger contraction produced faster or firmer movement. Small circuits built from integrated circuits replaced bulky electronics, and rechargeable nickel and lithium-ion batteries replaced earlier cells. Controllers could also switch between modes, such as elbow control and hand control, by using a quick co-contraction or a small switch.
Hands became more capable in the 2000s. A commercial hand with a separate motor for each finger appeared around 2007, and some models can be set to different grip patterns.
Targeted muscle reinnervation
A person who has lost an arm at the shoulder has few remaining muscles for control, and conventional myoelectric systems struggle to give enough independent signals. In the early 2000s, a surgeon, Todd Kuiken, and colleagues in Chicago developed targeted muscle reinnervation. In this surgery, nerves that once controlled the arm are redirected to grow into spare muscles, such as the chest. When the user thinks about moving the hand, the chest muscle contracts, and electrodes pick up the signal. The first patients reported in the early 2000s showed that intuitive control of several joints was possible. Some patients also reported that touching the reinnervated skin produced a sensation in the missing hand, which became a starting point for sensory feedback research.
DARPA and the Luke arm
In 2006, a United States defense research agency launched a program called Revolutionizing Prosthetics, aimed at creating advanced arms for injured service members. One result was an arm developed by a team led by the inventor Dean Kamen, which gave users many powered joints and a choice of control inputs, including foot sensors. The US Food and Drug Administration cleared it in 2014. It is an engineering achievement, but it is also expensive, and outcomes for individual users vary.
Limits that remain
Powered arms are not always preferred. They are heavier than body-powered ones, they need charging, and sweat and moisture can interfere with electrodes. Some users abandon them because of weight, cost or slow response, and many prefer a simpler hook for work. Surveys over the years show that rejection rates for upper-limb devices are higher than for leg devices, though the figures vary by study and are hard to compare.
Research into reading signals by pattern recognition, which uses artificial intelligence methods to interpret many electrodes at once, has made more natural control possible in laboratories. Moving it into daily use is slower. Future progress will depend not only on better electronics, but also on comfortable sockets, durable batteries and reliable touch feedback, for which some laboratories are testing artificial skin that senses pressure.