Online Learning Koopman Operator for Closed-Loop Electrical Neurostimulation in Epilepsy

Liang, Zhichao; Luo, Zixiang; Liu, Keyin; Qiu, Jingwei; Liu, Quanying

doi:10.1109/jbhi.2022.3210303

Cited by 10 publications

(2 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Extending our adaptive estimator to simultaneously estimate the linear matrix parameters online and completely remove the offline model fitting experiment is another direction that requires further investigation [71][72][73]. Third, there are rich explorations of effective feedback signals and closed-loop neuromodulation treatments for other neurological and neuropsychiatric disorders such as epilepsy [74][75][76] and major depression [39,77,78]. Our robust adaptive DBS control framework is flexible to model different types of input DBS parameters and output neural activity.…”

Section: Discussionmentioning

confidence: 99%

Robust adaptive deep brain stimulation control of in-silico non-stationary Parkinsonian neural oscillatory dynamics

Fang,

Berman,

Wang

et al. 2024

J. Neural Eng.

View full text Add to dashboard Cite

Objective. Closed-loop deep brain stimulation (DBS) is a promising therapy for Parkinson's disease (PD) that works by adjusting DBS patterns in real time from the guidance of feedback neural activity. Current closed-loop DBS mainly uses threshold-crossing on-off controllers or linear time-invariant (LTI) controllers to regulate the basal ganglia (BG) Parkinsonian beta band oscillation power. However, the critical cortex-BG-thalamus network dynamics underlying PD are nonlinear, non-stationary, and noisy, hindering accurate and robust control of Parkinsonian neural oscillatory dynamics. Approach. Here, we develop a new robust adaptive closed-loop DBS method for regulating the Parkinsonian beta oscillatory dynamics of the cortex-BG-thalamus network. We first build an adaptive state-space model to quantify the dynamic, nonlinear, and non-stationary neural activity. We then construct an adaptive estimator to track the nonlinearity and non-stationarity in real time. We next design a robust controller to automatically determine the DBS frequency based on the estimated Parkinsonian neural state while reducing the system's sensitivity to high-frequency noise. We adopt and tune a biophysical cortex-BG-thalamus network model as an in-silico simulation testbed to generate nonlinear and non-stationary Parkinsonian neural dynamics for evaluating DBS methods. Main results. We find that under different nonlinear and non-stationary neural dynamics, our robust adaptive DBS method achieved accurate regulation of the BG Parkinsonian beta band oscillation power with small control error, bias, and deviation. Moreover, the accurate regulation generalizes across different therapeutic targets and consistently outperforms current on-off and LTI DBS methods. Significance. These results have implications for future designs of closed-loop DBS systems to treat PD.

show abstract

Section: Discussionmentioning

confidence: 99%

Robust adaptive deep brain stimulation control of in-silico non-stationary Parkinsonian neural oscillatory dynamics

Fang,

Berman,

Wang

et al. 2024

J. Neural Eng.

View full text Add to dashboard Cite

show abstract

“…Based on the current results, it is reasonable to speculate that modulation of occipital closed-loop pathways are a promising target for the purpose of improving patients' emotion and cognition impairment in MRI − -DRE (e.g., modulation of SOG FC using neurostimulation, stereotactic EEG, or targeted drugs). (Aicua Rapun et al, 2019; Liang et al, 2022;Wang et al, 2022) Changes in closed-loop pathways including the SOG plus either middle occipital gyrus (MOG) or lingual gyrus (LING) were also correlated with cognitive dysfunction. These ndings are consistent with previous studies reporting that cognitively impaired elderly with white matter hyperintensities exhibited reduced amplitude of low-frequency uctuations in the MOG and that abnormal thickness of the LING was associated with social cognitive de cits in myotonic dystrophy type 1 patients (Serra et al, 2020;Xing et al, 2022).…”

Section: Discussionmentioning

confidence: 99%

Identification of abnormal closed-loop pathways in patients with MRI-negative pharmacoresistant epilepsy

Bu,

Ren,

Wang

et al. 2024

Brain Imaging and Behavior

View full text Add to dashboard Cite

To identify abnormal changes in closed-loop pathways among magnetic resonance imaging-negative drug-resistant epilepsy (MRI − -DRE) patients and to examine the associations of these abnormalities with emotional and cognitive impairments. A total of 26 patients with MRI − -DRE and 26 healthy controls (HCs) were included in this study. Causal brain networks and temporal-lag brain networks were constructed from rs-fMRI data, and the Johnson algorithm was used to identify stable closed-loop pathways. Abnormal closed-loop pathways in the MRI − -DRE group were identi ed by comparison with HCs, and associations with indicators of cognitive and emotional impairments were examined using Pearson correlation analysis. The results reveal that the abnormal stable closed-loop pathways were distributed across frontal, parietal, and occipital lobes, and included altered FC values both within and between cerebral hemispheres. Four abnormal closed-loop pathways in the occipital lobe were associated with emotional and cognitive impairments. These abnormalities may serve as biomarkers for diagnosis and guidance of individualized treatments for MRI − -DRE.

show abstract

Modelling and Controlling System Dynamics of the Brain: An Intersection of Machine Learning and Control Theory

Liu,

Wei,

et al. 2024

Advances in Neurobiology

View full text Add to dashboard Cite

Online Learning Koopman Operator for Closed-Loop Electrical Neurostimulation in Epilepsy

Cited by 10 publications

References 45 publications

Robust adaptive deep brain stimulation control of in-silico non-stationary Parkinsonian neural oscillatory dynamics

Robust adaptive deep brain stimulation control of in-silico non-stationary Parkinsonian neural oscillatory dynamics

Identification of abnormal closed-loop pathways in patients with MRI-negative pharmacoresistant epilepsy

Modelling and Controlling System Dynamics of the Brain: An Intersection of Machine Learning and Control Theory

Contact Info

Product

Resources

About