The effect of deep brain stimulation in Parkinson’s disease reflected in EEG microstates

Abstract

Mechanisms of deep brain stimulation (DBS) on cortical networks were explored mainly by fMRI. Advanced analysis of high-density EEG is a source of additional information and may provide clinically useful biomarkers. The presented study evaluates EEG microstates in Parkinson’s disease and the effect of DBS of the subthalamic nucleus (STN). The association between revealed spatiotemporal dynamics of brain networks and changes in oscillatory activity and clinical examination were assessed. Thirty-seven patients with Parkinson’s disease treated by STN-DBS underwent two sessions (OFF and ON stimulation conditions) of resting-state EEG. EEG microstates were analyzed in patient recordings and in a matched healthy control dataset. Microstate parameters were then compared across groups and were correlated with clinical and neuropsychological scores. Of the five revealed microstates, two differed between Parkinson’s disease patients and healthy controls. Another microstate differed between ON and OFF stimulation conditions in the patient group and restored parameters in the ON stimulation state toward to healthy values. The mean beta power of that microstate was the highest in patients during the OFF stimulation condition and the lowest in healthy controls; sources were localized mainly in the supplementary motor area. Changes in microstate parameters correlated with UPDRS and neuropsychological scores. Disease specific alterations in the spatiotemporal dynamics of large-scale brain networks can be described by EEG microstates. The approach can reveal changes reflecting the effect of DBS on PD motor symptoms as well as changes probably related to non-motor symptoms not influenced by DBS.

Vertical topography in EEG microstates: Physiology or artifact manifestation?

“…The analysis of EEG microstates is a popular technique for exploring scalp EEG data with a focus on the behavior of large-scale brain networks and brain dynamics (Khanna et al, 2015;Michel and Koenig, 2018). The spatiotemporal parameters of the main prototypical microstate maps are evaluated in healthy subjects (Bréchet et al, 2019;Jabès et al, 2021;Liu et al, 2020), and their alterations are explored in the context of psychiatric (Nishida et al, 2013;Sverak et al, 2021) and neurological diseases (Lamoš et al, 2023(Lamoš et al, , 2021. Some EEG/fMRI studies (Agrawal et al, 2022;Rajkumar et al, 2021;Schwab et al, 2015;Zappasodi et al, 2019) reported and assigned physiological significance to a topography with a distinct vertical line; we refer to this vertical topography as VT. Our study demonstrates that VT is partially a reflection of artifacts very likely caused by the movement of metallic parts of the EEG cap.…”

A longitudinal electrophysiological and behavior dataset for PD rat in response to deep brain stimulation

“…The generation, coding, transmission and integration of information in the brain are completed through the action potentials of neurons, because neuron is the smallest structure and functional unit of the nervous system 4 – 6 . The recording of individual neuron activity and local neuron population activity could help obtain the characteristics of action potentials, and also decode brain activity from multiple scales 7 – 9 , which is particularly important for exploring the mechanisms of DBS 10 , 11 .…”

Synaptic Insights: A Holistic Approach to Brain Pathology Classification through Transfer Learning and Ensemble Techniques

“…These algorithms aim to provide a non-invasive and cost-effective means of diagnosing PD. Brain conditions like tumors, Parkinson's disease, and multiple sclerosis can be detected through various brain imaging methods, including structural magnetic resonance imaging (sMRI), functional magnetic resonance imaging (fMRI), positron emission tomography (PET), or electroencephalography (EEG) [2]. Various machine learning and image processing techniques are employed to identify brain disorders.…”