EEG-based emotion charting for Parkinson's disease patients using Convolutional Recurrent Neural Networks and cross dataset learning

Multi-Scale Feature and Multi-Channel Selection toward Parkinson's Disease Diagnosis with EEG

OBJECTIVE

Motivated by Health Care 4.0, this study aims to reducing the dimensionality of traditional EEG features based on manual extracted features, including statistical features in the time and frequency domains.

METHODS

A total of 22 multi-scale features were extracted from the UNM and Iowa datasets using a 4th order Butterworth filter and wavelet packet transform. Based on single-channel validation, 29 channels with the highest R2 scores were selected from a pool of 59 common channels. The proposed channel selection scheme was validated on the UNM dataset and tested on the Iowa dataset to compare its generalizability against models trained without channel selection.

RESULTS

The experimental results demonstrate that the proposed model achieves an optimal classification accuracy of 100%. Additionally, the generalization capability of the channel selection method is validated through out-of-sample testing based on the Iowa dataset Conclusions: Using single-channel validation, we proposed a channel selection scheme based on traditional statistical features, resulting in a selection of 29 channels. This scheme significantly reduced the dimensionality of EEG feature vectors related to Parkinson's disease by 50%. Remarkably, this approach demonstrated considerable classification performance on both the UNM and Iowa datasets. For the closed-eye state, the highest classification accuracy achieved was 100%, while for the open-eye state, the highest accuracy reached 93.75%.

Parkinson's disease diagnosis using deep learning: A bibliometric analysis and literature review

Parkinson's Disease (PD) is a progressive neurodegenerative illness triggered by decreased dopamine secretion. Deep Learning (DL) has gained substantial attention in PD diagnosis research, with an increase in the number of published papers in this discipline. PD detection using DL has presented more promising outcomes as compared with common machine learning approaches. This article aims to conduct a bibliometric analysis and a literature review focusing on the prominent developments taking place in this area. To achieve the target of the study, we retrieved and analyzed the available research papers in the Scopus database. Following that, we conducted a bibliometric analysis to inspect the structure of keywords, authors, and countries in the surveyed studies by providing visual representations of the bibliometric data using VOSviewer software. The study also provides an in-depth review of the literature focusing on different indicators of PD, deployed approaches, and performance metrics. The outcomes indicate the firm development of PD diagnosis using DL approaches over time and a large diversity of studies worldwide. Additionally, the literature review presented a research gap in DL approaches related to incremental learning, particularly in relation to big data analysis.

Joint domain symmetry and predictive balance for cross-dataset EEG emotion recognition

BACKGROUND

Cross-dataset EEG emotion recognition is an extremely challenging task, since data distributions of EEG from different datasets are greatly different, which makes the universal models yield unsatisfactory results. Although there are many methods have been proposed to reduce cross-dataset distribution discrepancies, they still neglected the following two problems. (1) Label space inconsistency: emotional label spaces of subjects from different datasets are different; (2) Uncertainty propagation: the uncertainty of misclassified emotion samples will propagate between datasets.

NEW METHOD

To solve these problems, we propose a novel method called domain symmetry and predictive balance (DSPB). For the problem of label space inconsistency, a domain symmetry module is designed to make label spaces of source and target domain to be the same, which randomly selects samples from the source domain and put into the target domain. For the problem of uncertainty propagation, a predictive balance module is proposed to reduce the prediction score of incorrect samples and then effectively reduce distribution differences between EEG from different datasets.

RESULTS

Experimental results show that our method achieve 61.48% average accuracies on the three cross-dataset tasks. Moreover, we find that gamma is the most relevant to emotion recognition among the five frequency bands, and the prefrontal and temporal brain regions are the channels carrying the most emotional information among the 62 brain channels.

COMPARISON WITH EXISTING METHODS

Compared with the partial domain adaptation method (SPDA) and the unsupervised domain adaptation (MS-MDA), our method improves average accuracies by 15.60% and 23.11%, respectively.

CONCLUSION

Besides, data distributions of EEG from different datasets but with the same emotional labels have been well aligned, which demonstrates the effectiveness of DSPB.

Graph Neural Networks for Parkinson's Disease Monitoring and Alerting

Graph neural networks (GNNs) have been increasingly employed in the field of Parkinson's disease (PD) research. The use of GNNs provides a promising approach to address the complex relationship between various clinical and non-clinical factors that contribute to the progression of PD. This review paper aims to provide a comprehensive overview of the state-of-the-art research that is using GNNs for PD. It presents PD and the motivation behind using GNNs in this field. Background knowledge on the topic is also presented. Our research methodology is based on PRISMA, presenting a comprehensive overview of the current solutions using GNNs for PD, including the various types of GNNs employed and the results obtained. In addition, we discuss open issues and challenges that highlight the limitations of current GNN-based approaches and identify potential paths for future research. Finally, a new approach proposed in this paper presents the integration of new tasks for the engineering of GNNs for PD monitoring and alert solutions.

Entropy-Based Machine Learning Model for Fast Diagnosis and Monitoring of Parkinson's Disease

This study presents the concept of a computationally efficient machine learning (ML) model for diagnosing and monitoring Parkinson's disease (PD) using rest-state EEG signals (rs-EEG) from 20 PD subjects and 20 normal control (NC) subjects at a sampling rate of 128 Hz. Based on the comparative analysis of the effectiveness of entropy calculation methods, fuzzy entropy showed the best results in diagnosing and monitoring PD using rs-EEG, with classification accuracy (ARKF) of ~99.9%. The most important frequency range of rs-EEG for PD-based diagnostics lies in the range of 0-4 Hz, and the most informative signals were mainly received from the right hemisphere of the head. It was also found that ARKF significantly decreased as the length of rs-EEG segments decreased from 1000 to 150 samples. Using a procedure for selecting the most informative features, it was possible to reduce the computational costs of classification by 11 times, while maintaining an ARKF ~99.9%. The proposed method can be used in the healthcare internet of things (H-IoT), where low-performance edge devices can implement ML sensors to enhance human resilience to PD.

Diagnosis of Parkinson's disease based on voice signals using SHAP and hard voting ensemble method

Parkinson's disease (PD) is the second most common progressive neurological condition after Alzheimer's. The significant number of individuals afflicted with this illness makes it essential to develop a method to diagnose the conditions in their early phases. PD is typically identified from motor symptoms or via other Neuroimaging techniques. Expensive, time-consuming, and unavailable to the general public, these methods are not very accurate. Another issue to be addressed is the black-box nature of machine learning methods that needs interpretation. These issues encourage us to develop a novel technique using Shapley additive explanations (SHAP) and Hard Voting Ensemble Method based on voice signals to diagnose PD more accurately. Another purpose of this study is to interpret the output of the model and determine the most important features in diagnosing PD. The present article uses Pearson Correlation Coefficients to understand the relationship between input features and the output. Input features with high correlation are selected and then classified by the Extreme Gradient Boosting, Light Gradient Boosting Machine, Gradient Boosting, and Bagging. Moreover, the weights in Hard Voting Ensemble Method are determined based on the performance of the mentioned classifiers. At the final stage, it uses SHAP to determine the most important features in PD diagnosis. The effectiveness of the proposed method is validated using 'Parkinson Dataset with Replicated Acoustic Features' from the UCI machine learning repository. It has achieved an accuracy of 85.42%. The findings demonstrate that the proposed method outperformed state-of-the-art approaches and can assist physicians in diagnosing Parkinson's cases.

Diagnosis and classification of Parkinson's disease using ensemble learning and 1D-PDCovNN

In this paper, we proposed a novel approach to diagnose and classify Parkinson's Disease (PD) using ensemble learning and 1D-PDCovNN, a novel deep learning technique. PD is a neurodegenerative disorder; early detection and correct classification are essential for better disease management. The primary aim of this study is to develop a robust approach to diagnosing and classifying PD using EEG signals. As the dataset, we have used the San Diego Resting State EEG dataset to evaluate our proposed method. The proposed method mainly consists of three stages. In the first stage, the Independent Component Analysis (ICA) method has been used as the pre-processing method to filter out the blink noises from the EEG signals. Also, the effect of the band showing motor cortex activity in the 7-30 Hz frequency band of EEG signals in diagnosing and classifying Parkinson's disease from EEG signals has been investigated. In the second stage, the Common Spatial Pattern (CSP) method has been used as the feature extraction to extract useful information from EEG signals. Finally, an ensemble learning approach, Dynamic Classifier Selection (DCS) in Modified Local Accuracy (MLA), has been employed in the third stage, consisting of seven different classifiers. As the classifier method, DCS in MLA, XGBoost, and 1D-PDCovNN classifier has been used to classify the EEG signals as the PD and healthy control (HC). We first used dynamic classifier selection to diagnose and classify Parkinson's disease (PD) from EEG signals, and promising results have been obtained. The performance of the proposed approach has been evaluated using the classification accuracy, F-1 score, kappa score, Jaccard score, ROC curve, recall, and precision values in the classification of PD with the proposed models. In the classification of PD, the combination of DCS in MLA achieved an accuracy of 99,31%. The results of this study demonstrate that the proposed approach can be used as a reliable tool for early diagnosis and classification of PD.

Leveraging Computational Intelligence Techniques for Diagnosing Degenerative Nerve Diseases: A Comprehensive Review, Open Challenges, and Future Research Directions

Degenerative nerve diseases such as Alzheimer's and Parkinson's diseases have always been a global issue of concern. Approximately 1/6th of the world's population suffers from these disorders, yet there are no definitive solutions to cure these diseases after the symptoms set in. The best way to treat these disorders is to detect them at an earlier stage. Many of these diseases are genetic; this enables machine learning algorithms to give inferences based on the patient's medical records and history. Machine learning algorithms such as deep neural networks are also critical for the early identification of degenerative nerve diseases. The significant applications of machine learning and deep learning in early diagnosis and establishing potential therapies for degenerative nerve diseases have motivated us to work on this review paper. Through this review, we covered various machine learning and deep learning algorithms and their application in the diagnosis of degenerative nerve diseases, such as Alzheimer's disease and Parkinson's disease. Furthermore, we also included the recent advancements in each of these models, which improved their capabilities for classifying degenerative nerve diseases. The limitations of each of these methods are also discussed. In the conclusion, we mention open research challenges and various alternative technologies, such as virtual reality and Big data analytics, which can be useful for the diagnosis of degenerative nerve diseases.

Parkinson's disease is characterized by sub-second resting-state spatio-oscillatory patterns: A contribution from deep convolutional neural network

Deep convolutional neural network (DCNN) provides a multivariate framework to detect relevant spatio-oscillatory patterns in the data beyond common mass-univariate statistics. Yet, its practical application is limited due to the low interpretability of the results beyond accuracy. We opted to use DCNN with a minimalistic architecture design and large penalized terms to yield a generalizable and clinically relevant network model. Our network was trained based on the scalp topology of the electroencephalography (EEG) from an open access dataset, constituting our primary sample of healthy controls (n = 25) and Parkinson's disease (PD) patients (n = 25), with and without medication. Next, we validated the model on another independent, yet comparable open access EEG dataset (healthy controls (n = 20) and PD patients (n = 20)), which was unseen to the network. We applied Gradient-weighted Class Activation Mapping (Grad-CAM) interpretability technique to create a localization map exhibiting the key network predictors, based on the gradients of the classification score flowing into the last convolutional layer. Accordingly, our results indicated that a sub-second of intrinsic oscillatory power pattern in the beta band over the occipitoparietal, gamma band over the left motor cortex as well as theta band over the frontoparietal cluster, had the largest impact on the network score for dissociating the PD patients from age- and gender-matched healthy controls, across the two datasets. We further found that the off-medication motor symptoms were related to the occipitoparietal off-medication beta power whereas the disease duration was associated with the off-medication beta power of the motor cortex. The on-medication theta power of the frontoparietal was related to the improvement of the motor symptoms. In conclusion, our method enabled us to characterize PD patho-electrophysiology according to the multivariate topographic analysis approach, where both spatial and frequency aspects of the oscillations were simultaneously considered. Moreover, our approach was free from common reference problem of the EEG data analyses.

NoLiTiA: An Open-Source Toolbox for Non-linear Time Series Analysis

In many scientific fields including neuroscience, climatology or physics, complex relationships can be described most parsimoniously by non-linear mechanics. Despite their relevance, many neuroscientists still apply linear estimates in order to evaluate complex interactions. This is partially due to the lack of a comprehensive compilation of non-linear methods. Available packages mostly specialize in only one aspect of non-linear time-series analysis and most often require some coding proficiency to use. Here, we introduce NoLiTiA, a free open-source MATLAB toolbox for non-linear time series analysis. In comparison to other currently available non-linear packages, NoLiTiA offers (1) an implementation of a broad range of classic and recently developed methods, (2) an implementation of newly proposed spatially and time-resolved recurrence amplitude analysis and (3) an intuitive environment accessible even to users with little coding experience due to a graphical user interface and batch-editor. The core methodology derives from three distinct fields of complex systems theory, including dynamical systems theory, recurrence quantification analysis and information theory. Besides established methodology including estimation of dynamic invariants like Lyapunov exponents and entropy-based measures, such as active information storage, we include recent developments of quantifying time-resolved aperiodic oscillations. In general, the toolbox will make non-linear methods accessible to the broad neuroscientific community engaged in time series processing.