Determining the optimal stimulus waveforms of deep brain stimulation based on support vector machine

摘要

Deep brain stimulation (DBS) has been proven to be an effective treatment for neurological diseases. Currently, selection of stimulation parameters of DBS is mainly relying on the trial and error method which is time-consuming and cannot obtain an optimal treatment performance. Thus, determining parameters of DBS based on the real time states of the brain is appealing. In this paper, a method for choosing parameters of DBS according to firing rates of neurons based on support vector machine is presented. Firstly, the effects of different DBS parameters on the firing rates of neurons are analyzed, then the mapping relationships from DBS inputs to the firing rates of neurons are classified by support vector machine (SVM). Thus the optimal stimulus waveforms of DBS which could result in expected firing rates of neurons can be searched based on the mapping relationships.