Adapting the listening time for micro-electrode recordings in deep brain stimulation interventions

Abstract

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

Machine learning in deep brain stimulation: A systematic review

“…The second most commonly investigated phase is the surgery itself, with the clinical problem frequently being the inter-operative identification of the DBS target. Every paper addressing this problem used MER analysis [10,14,21,28,33,37,39,59,63,68,70,72,76,77,78]. Instead of helping clinicians to aim for an anatomical structure, Lu et al [69] proposed a method to predict, by analyzing MER, whether or not the electrode lead is inside a clinically predefined therapeutic site of activation.…”

“…CNNs are also extensively used for MER spectrogram analysis: three times with a structure based on 1D separable convolutions [70,77,78], once with the AlexNet model [53], once with a CNN based on VGG16 and trained with multi-task learning [73], and once with a custom structure based on 1D-convolution [76]. RNNs were used once with LSTM for MER artifact detection [61].…”

“…However, there were some papers that simply did not separate training and testing patients, could lead to biased measurements of their performance. The results reported by these contributions are interesting as preliminary work, but cannot safely be considered as representative of a real, prospective usage.Others measured their performance by employing a patient-wise validation method[11,12,21,22,26,27,32,29,33,36,47,48,51,54,55,57,58,61,62,63,64,65,66,69,70,72,73,74,75,76,77,78,79,80,81,82]. While it does not ensure the complete absence of data leakage (that could occur, for example, by mixing the validation and testing sets, or by selecting or normalize features with the whole database), these results can be considered as more reliable, and more representative of prospective-usage performance.…”