From fixed settings to adjustment
A “smart” prosthetic is one that senses what the user is doing and changes its behavior. The most established example is the microprocessor-controlled knee. Before these, knees had mechanical or hydraulic resistance set once by a technician, who tuned it for a particular walking speed. If the user walked faster or slower, the leg did not adjust, and the user compensated with the body. The development of the smart knee is the story of making that adjustment automatic.
Before electronics
For much of the twentieth century, the main problem for an above-knee amputee was controlling a knee that had no muscle behind it. A knee that swings freely may buckle under load, while a knee that is locked leaves the user walking stiff-legged. Designers tried friction brakes, which resist swing at a constant level, and later hydraulic and pneumatic cylinders, which resist more at faster speeds so that the leg swings through at a natural pace over a range of walking speeds. Weight-activated stance control locked the knee under load and released it when weight shifted off. These mechanical systems were a real improvement, and they are still in wide use. Their limits were that they were tuned for steady, level walking.
Electronics enter the knee
In the early 1990s, a British company introduced a knee in which a small computer adjusted the swing-phase resistance in real time, based on how fast the user was walking. It is generally dated to around 1993 and is often called the first commercially available computer-controlled prosthetic knee. Because the details of early development are not widely documented, it is best to treat the date as approximate. It controlled the swing phase, the part of a step when the leg is off the ground.
A major step followed in 1997, when a German company released a knee that controlled both swing and stance. Sensors measured the knee angle and the load through the leg many times per second, and a microprocessor adjusted a hydraulic valve to change the resistance. The user could walk more smoothly, descend stairs and slopes with a controlled bend, and in some cases recover from a stumble because the knee could stiffen instantly under load. That design is often described as the beginning of the modern smart leg. Other manufacturers produced competing knees through the 2000s and 2010s, and the technology spread to several price levels.
What the electronics sense
Typical sensors include:
- knee angle, to know the phase of the step
- load or force in the shank, to know if the user is putting weight on the leg
- an accelerometer or gyroscope, to detect motion and orientation
The microprocessor reads these signals hundreds of times a second and compares them with stored patterns for walking, standing, sitting and stairs. The ability to do this in a small, battery-powered unit depended on advances in integrated circuits and silicon chips, and on lithium-ion batteries that could last for a day. Newer knees can be adjusted by a clinician via software on a laptop or tablet, and some can be tuned with a phone.
What changed for users
Research on microprocessor knees has produced generally positive but nuanced findings. Several studies reported fewer stumbles and falls, easier walking on slopes and stairs, and users who said they felt more confident. Others found that benefits depend on the person’s activity level and that differences in measures such as energy use were small or inconsistent. Because knees are expensive, funding by insurers and health systems has been a matter of debate in many countries, and decisions often depend on evidence about how much a particular user would benefit. As with any study of individuals, results from one group do not necessarily apply to another.
Smart feet, ankles and sockets
The same approach was extended to other parts. Some ankles adjust their angle on slopes, for example by tilting the foot up as the user goes uphill, using a small motor and sensors. Hands controlled by muscle signals are related, but they use signals from the body rather than reading the limb’s own motion. Researchers are also testing sockets with built-in sensors that measure pressure and fit, to flag when a socket is causing problems, and prostheses that connect to a phone for data on use, an approach that overlaps with connected sensors and smart implants in other medical devices. These are mainly research or early commercial features, and their value is still being assessed.
Active devices that add energy, such as powered ankles and knees, form a different group. The microprocessor knee mostly controls how a joint resists, not how it pushes.
Limits and trade-offs
Electronics bring costs. Smart knees weigh more than simple ones, need charging, can be affected by water and require servicing. If the battery runs flat, many models revert to a basic safe mode, which is stiffer. Manufacturers usually specify water-resistance and the user’s weight limits, and clinicians guide choices. The history of the smart prosthetic suggests a pattern that repeats: a clever sensor-driven mechanism solves a real problem for some users, becomes more widely available, and then reveals which problems still need work.