Objective: Here, we tried to apply dynamic causal model to differentiate the motor network of dystonia patients from that of normal controls through EEG recordings.

Background: Task-specific dystonias are primary focal dystonias characterized by excessive muscle contractions producing abnormal postures during selective motor activities that often involve highly skilled, repetitive movements (Torres-Russotto et al., 2008). Although, the pathophysiology of dystonia may involve basal ganglion, cerebellum, and sensori-motor system, the cortical network has not been fully investigated.

Method: Total 15 task-specific focal hand dystonia with control were collected. We recorded EEG through a Neuroscan amplifier while subjects performing non-symptomatic hand movement. Data was then preprocessed with time-frequency Morlet wavelet transform. Five sources located in MNI space, including bilateral M1, bilateral PM and SMA with 6 plausible models. For the DCM analysis, we used Bayesian model selection to identify the optimal models among the models being tested both at the individual and the group levels. Once the optimal model were obtained, the frequency-specific coupling parameters were tested.

Results: In Fig 1., we compared the the model with nonlinear intrinsic and extrinsic coupling and fully connected networks of task-specific dystonia with normal control upon the 6 plausible models; among them, model 5 was the best model. We further identified five frequency-specific network connections that are significantly different between the patients and the healthy.

Conclusion: Here, dynamic causal model, we found a specific network difference between task-specific dystonia and control in group levels. The current results serve as a possible neuro-biomarker in the diagnosis of task-specific dystonia.

« Back to 2019 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/dynamic-causal-modelling-of-cortical-activity-in-task-specific-dystonia/