Brain structures recognition using MER signals and medical images - application to brain deep stimulation surgery

Abstract

Deep brain stimulation (DBS) is the most suitable surgical procedure for patients with Parkinson disease whose symptoms are not well controlled by drug treatment, or who cannot tolerate the side-effects of medication. The success of this surgery procedure, depends on the correct location of the neurostimulator device over the basal ganglia area (i.e. thalamus and globus pallidus). This paper presents a software system DEEPBRAINnav as support tool in Parkinson surgery. This software serves as support to the neurological specialist in DBS surgery. DEEPBRAINnav is composed mainly by two applications. First, it allows to the medical specialist, online recognition of brain structures by the analysis of signals from microelectrode recordings (MER), as well as analyses offline databases allowing the inclusion of new trained classifiers for automatic identification of thalamus location. Secondly, the software includes an interactive application to perform the tracking of the microelectrode device, allowing to the neurosurgeon visualize the path and real location of the microelectrode device implant process. DEEPBRAINnav has been tested for Deep Brain Stimulation surgeries performed at the Institute for Epilepsy and Parkinson of the Eje Cafetero (Colombia). The results shows an accurate tracking of the microelectrode implanting process and achieving positive identifications of the Subthalamic Nucleus (STN) over to 85% using a naive Bayes classifier. Furthermore, develop applications for image and signals guide surgery procedures, allows to the neurologist obtain more support information of a surgery procedure given, with the aim to increase the robustness of the DBS surgery.