Pathological tremor prediction using surface electromyogram and acceleration: potential use in 'ON-OFF' demand driven deep brain stimulator design.

摘要: Objective. We present a proof of concept for novel method predicting the onset pathological tremor using non-invasively measured surface electromyogram (sEMG) and acceleration from tremor-affected extremities patients with Parkinson's disease (PD) essential (ET). Approach. The prediction algorithm uses set spectral (Fourier wavelet) nonlinear time series (entropy recurrence rate) parameters extracted recorded sEMG signals. Main results. resulting is shown to successfully predict all 91 trials in 4 PD ET patients. predictor achieves 100% sensitivity considered, along an overall accuracy 85.7% 80.2% trials. By Pearson's chi-square test, results are significantly differ random outcome. Significance. can be potentially used designing next generation non-invasive closed-loop predictive ON–OFF controllers deep brain stimulation (DBS), suppressing such Such system based on alternating ON OFF DBS periods, incoming being predicted during intervals when OFF, so as turn back ON. should few seconds before re-appears that patient tremor-free entire cycle interval maximized order minimize current injected battery usage.