Dataset Information

Machine learning prediction of motor response after deep brain stimulation in Parkinson's disease-proof of principle in a retrospective cohort.

ABSTRACT: Introduction:Despite careful patient selection for subthalamic nucleus deep brain stimulation (STN DBS), some Parkinson's disease patients show limited improvement of motor disability. Innovative predictive analysing methods hold potential to develop a tool for clinicians that reliably predicts individual postoperative motor response, by only regarding clinical preoperative variables. The main aim of preoperative prediction would be to improve preoperative patient counselling, expectation management, and postoperative patient satisfaction. Methods:We developed a machine learning logistic regression prediction model which generates probabilities for experiencing weak motor response one year after surgery. The model analyses preoperative variables and is trained on 89 patients using a five-fold cross-validation. Imaging and neurophysiology data are left out intentionally to ensure usability in the preoperative clinical practice. Weak responders (n = 30) were defined as patients who fail to show clinically relevant improvement on Unified Parkinson Disease Rating Scale II, III or IV. Results:The model predicts weak responders with an average area under the curve of the receiver operating characteristic of 0.79 (standard deviation: 0.08), a true positive rate of 0.80 and a false positive rate of 0.24, and a diagnostic accuracy of 78%. The reported influences of individual preoperative variables are useful for clinical interpretation of the model, but cannot been interpreted separately regardless of the other variables in the model. Conclusion:The model's diagnostic accuracy confirms the utility of machine learning based motor response prediction based on clinical preoperative variables. After reproduction and validation in a larger and prospective cohort, this prediction model holds potential to support clinicians during preoperative patient counseling.

SUBMITTER: Habets JGV

PROVIDER: S-EPMC7680051 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning prediction of motor response after deep brain stimulation in Parkinson's disease-proof of principle in a retrospective cohort.

Habets Jeroen G V JGV Janssen Marcus L F MLF Duits Annelien A AA Sijben Laura C J LCJ Mulders Anne E P AEP De Greef Bianca B Temel Yasin Y Kuijf Mark L ML Kubben Pieter L PL Herff Christian C

PeerJ 20201118

<h4>Introduction</h4>Despite careful patient selection for subthalamic nucleus deep brain stimulation (STN DBS), some Parkinson's disease patients show limited improvement of motor disability. Innovative predictive analysing methods hold potential to develop a tool for clinicians that reliably predicts individual postoperative motor response, by only regarding clinical preoperative variables. The main aim of preoperative prediction would be to improve preoperative patient counselling, expectatio ...[more]

PMID: 33240642

Similar Datasets

Project description:IntroductionTranscranial magnetic stimulation (TMS) mapping has become a critical tool for exploratory studies of the human corticomotor (M1) organization. Here, we propose to gather existing cutting-edge TMS-EMG and TMS-EEG approaches into a combined multi-dimensional TMS mapping that considers local and whole-brain excitability changes as well as state and time-specific changes in cortical activity. We applied this multi-dimensional TMS mapping approach to patients with Parkinson's disease (PD) with Deep brain stimulation (DBS) of the sub-thalamic nucleus (STN) ON and OFF. Our goal was to identifying one or several TMS mapping-derived markers that could provide unprecedent new insights onto the mechanisms of DBS in movement disorders.MethodsSix PD patients (1 female, mean age: 62.5 yo [59-65]) implanted with DBS-STN for 1 year, underwent a robotized sulcus-shaped TMS motor mapping to measure changes in muscle-specific corticomotor representations and a movement initiation task to probe state-dependent modulations of corticospinal excitability in the ON (using clinically relevant DBS parameters) and OFF DBS states. Cortical excitability and evoked dynamics of three cortical areas involved in the neural control of voluntary movements (M1, pre-supplementary motor area - preSMA and inferior frontal gyrus - IFG) were then mapped using TMS-EEG coupling in the ON and OFF state. Lastly, we investigated the timing and nature of the STN-to-M1 inputs using a paired pulse DBS-TMS-EEG protocol.ResultsIn our sample of patients, DBS appeared to induce fast within-area somatotopic re-arrangements of motor finger representations in M1, as revealed by mediolateral shifts of corticomuscle representations. STN-DBS improved reaction times while up-regulating corticospinal excitability, especially during endogenous motor preparation. Evoked dynamics revealed marked increases in inhibitory circuits in the IFG and M1 with DBS ON. Finally, inhibitory conditioning effects of STN single pulses on corticomotor activity were found at timings relevant for the activation of inhibitory GABAergic receptors (4 and 20 ms).ConclusionTaken together, these results suggest a predominant role of some markers in explaining beneficial DBS effects, such as a context-dependent modulation of corticospinal excitability and the recruitment of distinct inhibitory circuits, involving long-range projections from higher level motor centers and local GABAergic neuronal populations. These combined measures might help to identify discriminative features of DBS mechanisms towards deep clinical phenotyping of DBS effects in Parkinson's Disease and in other pathological conditions.

Dataset Information

Machine learning prediction of motor response after deep brain stimulation in Parkinson's disease-proof of principle in a retrospective cohort.

Publications

Machine learning prediction of motor response after deep brain stimulation in Parkinson's disease-proof of principle in a retrospective cohort.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets