Automatic Sleep Stage Classification Based on Subthalamic Local Field Potentials

Abstract

Deep brain stimulation (DBS) is an established treatment for patients with Parkinson’s disease (PD). Sleep disorders are common complications of PD and affected by subthalamic DBS treatment. To achieve more precise neuromodulation, chronic sleep monitoring and closed-loop DBS toward sleep-wake cycles could potentially be utilized. Local field potential (LFP) signals that are sensed by the DBS electrode could be processed as primary feedback signals. This is the first study to systematically investigate the sleep-stage classification based on LFPs in subthalamic nucleus (STN). With our newly developed recording and transmission system, STN-LFPs were collected from 12 PD patients during wakefulness and nocturnal polysomnography sleep monitoring at one month after DBS implantation. Automatic sleep-stage classification models were built with robust and interpretable machine learning methods (support vector machine and decision tree). The accuracy, sensitivity, selectivity, and specificity of the classification reached high values (above 90% at most measures) at group and individual levels. Features extracted in alpha (8–13 Hz), beta (13–35 Hz), and gamma (35–50 Hz) bands were found to contribute the most to the classification. These results will directly guide the engineering development of implantable sleep monitoring and closed-loop DBS and pave the way for a better understanding of the STN-LFP sleep patterns.

Case Report: Lateral Habenula Deep Brain Stimulation for Treatment-Resistant Depression

“…All the LFP signals were sensed differentially through adjacency contacts by implantable pulse generator (IPG) with 1,000 Hz sampling rate. The detailed methods of the LFPs data collection and processing were described in our previous studies ( 14 ). The 10-min data was collected when the stimulation was off at each time.…”

Proceedings of the Eighth Annual Deep Brain Stimulation Think Tank: Advances in Optogenetics, Ethical Issues Affecting DBS Research, Neuromodulatory Approaches for Depression, Adaptive Neurostimulation, and Emerging DBS Technologies

“…In addition to this change we also observed combinations of alpha, beta and gamma bands which could be used as chronic biomarkers for the classification of different sleep stages ( Chen et al, 2019 ). The results have guided the development of a closed-loop DBS approach.…”

Deep Brain Stimulation Initiative: Toward Innovative Technology, New Disease Indications, and Approaches to Current and Future Clinical Challenges in Neuromodulation Therapy

“…The first clinical trial has been completed in PD patients ( n = 13) with a successful post-surgery follow-up through 12 months. Preliminary data have been used to characterize components of β-band oscillations during sleep states ( 11 ). Furthermore, it is reported that the response of β-band oscillations to high-frequency DBS is changed over time, which is a likely result from changes in neural network plasticity.…”

“…Future developments toward robust online learning techniques that explore large decision spaces and adapt to feedback in real time are essential to online learning problems. The application of AI techniques may allow us to uncover the mechanisms of DBS and the understanding of how DBS influences brain networks ( 11 ). It is noteworthy that recent advances in MRI-compatible DBS devices are allowing for acquisition of neuroimages during stimulation.…”