Background: Myoelectric controlled prosthetic hand requires machine based\nidentification of hand gestures using surface electromyogram (sEMG) recorded from\nthe forearm muscles. This study has observed that a sub-set of the hand gestures\nhave to be selected for an accurate automated hand gesture recognition, and reports a\nmethod to select these gestures to maximize the sensitivity and specificity.\nMethods: Experiments were conducted where sEMG was recorded from the muscles\nof the forearm while subjects performed hand gestures and then was classified off-line.\nThe performances of ten gestures were ranked using the proposed Positiveââ?¬â??Negative\nPerformance Measurement Index (PNM), generated by a series of confusion matrices.\nResults: When using all the ten gestures, the sensitivity and specificity was 80.0% and\n97.8%. After ranking the gestures using the PNM, six gestures were selected and these\ngave sensitivity and specificity greater than 95% (96.5% and 99.3%); Hand open, Hand\nclose, Little finger flexion, Ring finger flexion, Middle finger flexion and Thumb flexion.\nConclusion: This work has shown that reliable myoelectric based human computer\ninterface systems require careful selection of the gestures that have to be recognized\nand without such selection, the reliability is poor
Loading....