Data Mining Tools and deep brain stimulation target evaluation

Parkinson´s disease is a slow, but inexorably progressive disorder; with unknown origin (only about 10% cases are genetically based). It is a chronic neurological disease, defined by tremor, slow motion and stiffness. The neurons do not produce dopamine anymore or produce very low level of this chemical mediator, necessary on movement coordination. Data testifies that in our country more than 90,000 patients suffer from this disease. The DBS (Deep Brain Stimulation) is an invasive method of applying electrical stimulus to a sum of thalamic nucleus at frequencies greater than 100Hz and represents now a successful method for reducing the tremor. First, we assessed the relevance of nonlinear dynamic parameters in tremor analysis. Second, we applied Data Mining Tools to solve the well known clustering and classification problem and DBS treatments results identify. In most cases, there is no way to predict or prevent sporadic PD. Being an invasive method it is important to predict patient´s response to DBS and therefore the next step is to determine a model of the Parkinson tremor.