Micro-electrode recording (MER) is a powerful way of localizing target structures during neurosurgical procedures such as the implantation of deep brain stimulation electrodes, which is a common treatment for Parkinson's disease and other neurological disorders. While Micro-electrode Recording (MER) provides adjunctive information to guidance assisted by pre-operative imaging, it is not unanimously used in the operating room. The lack of standard use of MER may be in part due to its long duration, which can lead to complications during the operation, or due to high degree of expertise required for their interpretation. Over the past decade, various approaches addressing automating MER analysis for target localization have been proposed, which have mainly focused on feature engineering. While the accuracies obtained are acceptable in certain configurations, one issue with handcrafted MER features is that they do not necessarily capture more subtle differences in MER that could be detected auditorily by an expert neurophysiologist. In this paper, we propose and validate a deep learning-based pipeline for subthalamic nucleus (STN) localization with micro-electrode recordings motivated by the human auditory system. Our proposed Convolutional Neural Network (CNN), referred as SepaConvNet, shows improved accuracy over two comparative networks for locating the STN from one second MER samples.

Deep brain stimulation (DBS) is an interventional treatment for Parkinson's disease which involves the precise positioning of stimulated electrodes within deep brain structures, such as the SubThalamic Nucleus (STN). Although originally identified via imaging, additional inter-operative guidance is necessary to localize the target anatomy. Analysis of Micro-Electrode Recordings (MERs) allows for a trained neurophysiologist to infer the underlying anatomy at a particular electrode position using human audition, although it is subjective and requires a high degree of expertise. Various approaches to assist MER analysis during DBS are proposed in the literature, including deep learning methods, which rely on a static input description, that is, a pre-defined number of features or input size. In this paper, we propose two dynamic deep learning approaches adaptable to the complexity of MERs signal, by using an arbitrary long listening time (in 1s chunks), while providing feedback to the neurophysiologist as to the model's certainty. We evaluated five different deep learning based classifiers which can use arbitrary length MERs for STN segmentation. We found that a Bayesian extension using the highlevel features from SepaConvNet performed the best, increasing the balanced accuracy to 83.5%. This work represents a step forward in integrating automated analysis of MERs into the DBS surgical workflow by automatically finding and exploiting possible efficiencies in MER acquisition.

Deep Brain Stimulation (DBS) is a common treatment for a variety of neurological disorders which involves the precise placement of electrodes at particular subcortical locations such as the subthalamic nucleus. This placement is often guided by auditory analysis of micro-electrode recordings (MERs) which informs the clinical team as to the anatomic region in which the electrode is currently positioned. Recent automation attempts have lacked flexibility in terms of the amount of signal recorded, not allowing them to collect more signal when higher certainty is needed or less when the anatomy is unambiguous. Methods: We have addressed this problem by evaluating a simple algorithm that allows for MER signal collection to terminate once the underlying model has sufficient confidence. We have parameterized this approach and explored its performance using three underlying models composed of one neural network and two Bayesian extensions of said network. Results: We have shown that one particular configuration, a Bayesian model of the underlying network's certainty, outperforms the others and is relatively insensitive to parameterization. Further investigation shows that this model also allows for signals to be classified earlier without increasing the error rate. Conclusions: We have presented a simple algorithm that records the confidence of an underlying neural network, thus allowing for MER data collection