EEG coherence as a diagnostic tool to measure the initial stages of Parkinson Disease

Evaluation of Parkinson's disease early diagnosis using single-channel EEG features and auditory cognitive assessment

Background

Parkinson's disease (PD) often presents with subtle early signs, making diagnosis difficult. F-DOPA PET imaging provides a reliable measure of dopaminergic function and is a primary tool for early PD diagnosis. This study aims to evaluate the ability of machine-learning (ML) extracted EEG features to predict F-DOPA results and distinguish between PD and non-PD patients. These features, extracted using a single-channel EEG during an auditory cognitive assessment, include EEG feature A0 associated with cognitive load in healthy subjects, and EEG feature L1 associated with cognitive task differentiation.

Methods

Participants in this study are comprised of cognitively healthy patients who had undergone an F-DOPA PET scan as a part of their standard care (n = 32), and cognitively healthy controls (n = 20). EEG data collected using the Neurosteer system during an auditory cognitive task, was decomposed using wavelet-packet analysis and machine learning methods for feature extraction. These features were used in a connectivity analysis that was applied in a similar manner to fMRI connectivity. A preliminary model that relies on the features and their connectivity was used to predict initially unrevealed F-DOPA test results. Then, generalized linear mixed models (LMM) were used to discern between PD and non-PD subjects based on EEG variables.

Results

The prediction model correctly classified patients with unrevealed scores as positive F-DOPA. EEG feature A0 and the Delta band revealed distinct activity patterns separating between study groups, with controls displaying higher activity than PD patients. In controls, EEG feature L1 showed variations between resting state and high-cognitive load, an effect lacking in PD patients.

Conclusion

Our findings exhibit the potential of single-channel EEG technology in combination with an auditory cognitive assessment to distinguish positive from negative F-DOPA PET scores. This approach shows promise for early PD diagnosis. Additional studies are needed to further verify the utility of this tool as a potential biomarker for PD.

Automated Parkinson's disease recognition based on statistical pooling method using acoustic features

Parkinson's disease is one of the mostly seen neurological disease. It affects to nervous system and hinders people's vital activities. The majority of Parkinson's patients lose their ability to speak, write and balance. Many machine learning methods have been proposed to automatically diagnose Parkinson's disease using acoustic, hand writing and gaits. In this study, a statistical pooling method is proposed to recognize Parkinson's disease using the vowels. The used Parkinson's disease dataset contains the features of vowels. In the proposed method, the features of dataset are increased by applying statistical pooling method. Then, the most weighted features are selected from increased feature vector by using ReliefF. The classification is applied using the most weighted feature vector obtained. In the proposed method, Support Vector Machine (SVM) and K Nearest Neighbor (KNN) algorithms are used. The success rate was calculated as 91.25% and 91.23% with by using SVM and KNN respectively. The proposed method has two main contributions. The first is to obtain new features from the Parkinson's acoustic dataset using the statistical pooling method. The second one is the selection of the most significant features from the many feature vectors obtained. Thus, successful results were obtained for both KNN and SVM algorithms. The comparatively results clearly show that the proposed method achieved the best success rate among the selected state-of-art methods. Considering the proposed method and the results obtained, it proposed method is successful for Parkinson's disease recognition.