A robotic elbow brace that senses the intention of its user and aids movement should soon be available to help stroke survivors perform everyday tasks, such as turning off light switches. What’s more, early trials suggest the device acts as a therapeutic aid, markedly improving a patient’s ability to move even without the device.

The robotic aid “will make rehab easier and far more effective,” says Rutledge Ellis-Behnke, an MIT neuroscientist. “In addition, it will enable patients to work as much as they want for as long as they want. It enables additional rehab outside of the rehab hospital.”

Mira Sahney, president and one of the founders of Myomo, the Boston-based company that has been developing the brace, says they expect to have FDA approval, which requires certifying the safety of the device, within six weeks. Meanwhile, the robotic brace will be available for use in some New England clinics. By the end of this year or the beginning of 2007, depending on funding, the $6,000 machines should be available for general purchase and home use, she says. The company’s next steps include adapting the device for aiding movement of the wrist and hand, in addition to the elbow.

The “active” brace was invented in 2003 at MIT by engineering graduate students in a group led by Woodie Flowers, MIT professor of mechanical engineering. John McBean and Kailas Narendran, also founders of Myomo, used myoelectric sensors on the skin to detect faint voltage changes in underlying muscles as users attempted to move their arms. Since the signals “from someone who is essentially paralyzed are very, very small,” Sahney says, software and electronics are needed to filter out background noise and boost the signal to direct an electric motor to bend or extend the arm. Often patients can move their arms on their own to a certain point, but no further – at this point the signal from the muscles is interrupted. To fix this problem, the engineers have written software that helps “smooth out the signal,” filling in these interruptions in signal and allowing the motor to keep moving the patient’s arm. “A lot of times they’ll have dead areas, and so we try to help them get over that area,” she says.

Physical therapists can adjust the robotic brace to provide just enough assistance to patients, so they use their own muscles as much as possible, preventing atrophy. The active involvement of the patient has another advantage: it may help promote nerve-cell growth, reestablishing connections damaged in a stroke, and helping patients move more on their own.

In a pilot trial of the device, six stroke survivors who had shown no improvement in their ability to move for at least six months improved by an average 30 percent on one physical therapy scale of arm function. For example, often with stroke, muscles involuntary tighten, causing a patient’s arm to curl up and be difficult to extend. All of the patients were able to extend their arm to a greater degree after training with the robot. “It was beyond our dreams of what we might be able to accomplish,” Sahney says. Future studies will feature more patients and rigorous controls.

Other robotic devices for helping stroke patients are being developed and tested, including a walking assistant by Chicago PT of Evanston, IL, and a range of rehabilitative robots for exercising arms, wrists, and legs by Interactive Motion Technologies of Cambridge, MA.