| Object: The goal in fitting an upper limb amputee with arm prosthesis is to restore the lost function and appearance of the missing limb. Though the prevailing myoelectrically controlled prosthetic hand can restore appearance and some functions of the missing hand, the accuracy of motion is not satisfying and affected by many factors. It is widely accepted by most researchers that the electronic prosthetic hand is only practical when the correct recognition is greater than 95%. But motion accuracy of a myoelectrically controlled prosthetic hand was 45-90% (average 70%) because of internal defect of myoelectric signals. In order to improve the motion accuracy of a multiple freedom and function prosthesis, new signal sources must be found. The purpose of this experiment was to record signals of peripheral fasciculi with LIFEs and to discuss the possibility of these signals as a new signal source to control an electronic prosthetic hand.Method: LIFEs were made of Teflon-insulated 95% platinum 5% iridium wires with diameters of 60μm. Compared with intrafascicular needle electrodes made by Bionic Corporation, the recording and stimulating characters of LIFEs were detected by recording NAPs and CMAP when they were implanted into sciatic peripheral fasciculi of rabbits. Then biocompatibility of LIFEs with peripheral fasciculi was studied by implanting LIFEs chronically (six months) into sciatic peripheral fasciculi of 12 rabbits. In order to observe the change of signals recorded by LIFEs over six-month period, MEPs and CSEPs were detected respectively at 0, 0.5, 1, 2, 3, 4, 6 month after operation. At the end of the experiment, the fasciculi were anatomized for pathological examination. After it was proved that LIFEs had excellent biocompatibility with peripheral fasciculi by animal experiments, LIFEs were implanted into radial nerve, ulnar nerve and medial nerve of an amputee volunteer. Signals were detected when the volunteer was asked to do different movements of his missing hand, and the signals recorded by LIFEs were used to control an electronic prosthetic hand.Results: When LIFEs were used as recording electrodes, the average amplitudesof downward and upward NAPs recorded by LIFEs were respectively 550.0±92.0μV and 470.5±122.1μV. The average amplitudes of downward and upward NAPs recorded by intrafascicular needle electrodes were respectively 496.5±93.8μV and 425.0±76.7μV. The difference of average amplitudes of NAPs recorded by these two kinds of electrodes has no statistical significance (paired-samples t test, p>0.05). When LIFEs were used as stimulating electrodes, they have more selective than extraneural hook electrodes. When LIFEs were implanted chronically into sciatic fasciculi of rabbits for six months, 3/12 rabbits' LIFEs lost their recording abilities as they ruptured at 2, 3 and 6 month respectively after operation. When LIFEs were used as recording electrodes, results show the difference of OL and IPA of MEPs recorded by LIFEs implanted into different fasciculi of same rabbits at the same time has no statistical significance (paired-samples t test, p>0.05). But the difference of OL and IPA of MEPs recorded by LIFEs implanted into same fasciculus at different time has remarkable statistical significance (one way ANOVA, p<0.001). After multiple comparisons, we found OL of MEPs had no obvious change during 1 month, but significantly increased in the later time, and then became stable after 3 months. IPA of MEPs had no distinct change during 1 month, but significantly decreased over the next period, and then became stable after 3 months. When two different fasciculi were stimulated respectively through LIFEs, the difference of OL and IPA of CSEPs recorded by needle electrodes inserted into subcutaneous tissue of cranium at the same time has no statistical significance (paired-samples t test, p>0.05). When the same fasciculus was stimulated through LIFE, the difference of OL of CSEPs recorded by needle electrode at different time has remarkable statistical significance (one way ANO...
|
PDF Full Text Request