Desenvolvimento de um sistema de analise digital de sinais eletromiograficos

AUTOR(ES)
DATA DE PUBLICAÇÃO

1993

RESUMO

In order to implement neuromuscular stimulation control methods for rehabilitation of paraplegic and tetraplegic patients, the eletromyographic control arises as a viable proposal. However, there is the need for recognizing the muscular contraction patterns that happen just before the step performing. In order to extract parameters of the surface electromyographic signal and, from these parameters, find the desired patterns, a digital processing system for these signals was developed. The system acquires the electromyographic signals on the skin surface in various instances during the gait cycle and stores them in floppy disks. The system also reads this data and processes it, using user-specified algorithms. In order to make these functions possible, the system consists of an analogic signal conditioner, an Analogic-to-Digital converter, a microcomputer and a software which was developed to control the whole system. The initial algorithms which were used for the signal processing were : Autoregressive (AR) Model parameters, signal variance and FFT. Two patients were analised with this system. The brachial triceps muscle showed to be very promised. In order to evaluatethe feasibility of the electromyographic control, a neural network was used in trying to recognize the patterns. For these tests, the signal variance and the order 4 AR parameters were extracted. For one of the patients, the network did recognize the step-performing intention of the patient. For the other patient, the fault seems to be a consequence of a innervation defficiency of the analized muscle. The system seems to be very useful for gait electromyographic control studies, finding applications in the control of electromyographic prostheses. Besides that, it was noticedthe existence of a basic pattern during the brachial triceps contraction, which was different to that of the patient with innervation defficiency. As such, it seems that the system could find applications also in the clinical area, helping in the neuromuscular diseases diagnose. Thus, the system is a very useful and flexibletool,that can have other signal processing algorithms easily implemented

ASSUNTO(S)

eletromiografia engenharia biomedica

ACESSO AO ARTIGO

Documentos Relacionados