Several devices were proposed recently such as a device called Rewalk (demonstrated in Dallas) and a leg-rehabilitation device created by Timothy J Becker and Matthew Duplin. Also, an invasive technique in which a stent-based electrode called stentrode is implanted in the blood vessel was proposed. This technique gives patients mobility by just the power of thoughts. However, all these methods have a common drawback in that they require the paralyzed person to wear an exoskeleton which is a cumbersome.
We propose a noninvasive model called the Muscle to Machine Interface for Paralyzed person (MMIP) that does not require an exoskeleton for motion. Muscle electrodes that excite the muscle nerves are used. This system has the provision for preprocessing, classifying, recording, and training multidimensional EMG signals. The controlled person (paralyzed person) is interfaced with the movable part of the controlling person through the MMIP. When the controlling person does some operation, the MMIP generates electrical stimulation and in response to the stimulus the controlled person mimics the operation. A pair of EMG electrodes has to be placed on specific locations. The MMIP records the EMG signals from the controlled person, processes it and classifies it into different operations. The recorded operation is compared with the actual operations. The performance accuracy is determined as the number of correct classifications divided by number of physical operations. The phases are compared and tested for the accuracy. If the result is not satisfactory a bio-feedback is given and the process is repeated till the performance accuracy is achieved.
The system provides an easy to use non-invasive technique to improve the mobility of paralyzed people to help them lead a dignified life.
The block diagram flow is depicted in Figure 1.