Most of the traditional alcoholism detection methods are developed based on machine learning based methods that cannot extract the deep concealed characteristics of Electroencephalogram (EEG) signals from different layers. Hence, this study aims to introduce a deep leaning-based method that can automatically identify alcoholic EEG signals. It also explores if a hand-crafted feature extraction method is worth applying to deep learning techniques for classification of alcoholism. To investigate this, this paper presents two deep learning-based algorithms for classification of alcoholic EEG signals for comparison. In Algorithm 1, Principal Component Analysis (PCA) based feature extraction technique has been applied to extract representative components and then the extracted features are used as input to Artificial neural network (ANN) for classification. In Algorithm 2, the raw EEG data are directly used as inputs to a deep learning method: ‘long short-term memory (LSTM)’ for detection of alcoholism. The proposed algorithms were tested on a publicly available UCI Alcoholic EEG dataset. The experimental results show that the proposed Algorithm 2 could achieve an average classification accuracy of 93% while this accuracy is 86% for the proposed Algorithm 1. The comparative evaluations with the state-of-the-art algorithms indicate that Algorithm 2 also outperforms other competing algorithms in the literature. Thus deep learning algorithm when applied to raw data, can produce better performance than the combination of the hand-crafted feature method and the deep leaning algorithm. Our proposed system can be used to determine the extent of alcoholism-related changes in EEG signals and the effectiveness of therapeutic plans.
25 December 2023 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Mild Cognitive Impairment (MCI) is often considered a precursor to Alzheimer’s disease (AD), with a high likelihood of progression. Accurate and timely diagnosis of MCI is essential for halting the progression of AD and other forms of dementia. Electroencephalography (EEG) is the prevalent method for identifying MCI biomarkers. Frequency band-based EEG biomarkers are crucial for identifying MCI as they capture neuronal activities and connectivity patterns linked to cognitive functions. However, traditional approaches struggle to identify precise frequency band-based biomarkers for MCI diagnosis. To address this challenge, a novel framework has been developed for identifying important frequency sub-bands within EEG signals for MCI detection. In the proposed scheme, the signals are first denoised using a stationary wavelet transformation and segmented into small time frames. Then, four frequency sub-bands are extracted from each segment, and spectrogram images are generated for each sub-band as well as for the full filtered frequency band signal segments. This process produces five different sets of images for five separate frequency bands. Afterwards, a convolutional neural network is used individually on those image sets to perform the classification task. Finally, the obtained results for the tested four sub-bands are compared with the results obtained using the full bandwidth. Our proposed framework was tested on two MCI datasets, and the results indicate that the 16-32 Hz sub-band range has the greatest impact on MCI detection, followed by 4-8 Hz. Furthermore, our framework, utilizing the full frequency band, outperformed existing state-of-the-art methods, indicating its potential for developing diagnostic tools for MCI detection.
1 January 2018 in International Journal of Computational Intelligence Systems
This paper presents a computer aided analysis system for detecting epileptic seizure from electroencephalogram (EEG) signal data. As EEG recordings contain a vast amount of data, which is heterogeneous with respect to a time-period, we intend to introduce a clustering technique to discover different groups of data according to similarities or dissimilarities among the patterns. In the proposed methodology, we use K-means clustering for partitioning each category EEG data set (e.g. healthy; epileptic seizure) into several clusters and then extract some representative characteristics from each cluster. Subsequently, we integrate all the features from all the clusters in one feature set and then evaluate that feature set by three well-known machine learning methods: Support Vector Machine (SVM), Naive bayes and Logistic regression. The proposed method is tested by a publicly available benchmark database: 'Epileptic EEG database'. The experimental results show that the proposed scheme with SVM classifier yields overall accuracy of 100% for classifying healthy vs epileptic seizure signals and outperforms all the recent reported existing methods in the literature. The major finding of this research is that the proposed K-means clustering based approach has an ability to efficiently handle EEG data for the detection of epileptic seizure.
22 October 2016 in Environment, Development and Sustainability
Purpose: Mental health issues of young minds are at the threshold of all development and possibilities. Obsessive–compulsive disorder (OCD), separation anxiety disorder (SAD), and attention deficit hyperactivity disorder (ADHD) are three of the most common mental illness affecting children and adolescents. Several studies have been conducted on approaches for recognising OCD, SAD and ADHD, but their accuracy is inadequate due to limited features and participants. Therefore, the purpose of this study is to investigate the approach using machine learning (ML) algorithms with 1474 features from Australia's nationally representative mental health survey of children and adolescents. Methods: Based on the internal cross-validation (CV) score of the Tree-based Pipeline Optimization Tool (TPOTClassifier), the dataset has been examined using three of the most optimal algorithms, including Random Forest (RF), Decision Tree (DT), and Gaussian Naïve Bayes (GaussianNB). Results: GaussianNB performs well in classifying OCD with 91% accuracy, 76% precision, and 96% specificity as well as in detecting SAD with 79% accuracy, 62% precision, 91% specificity. RF outperformed all other methods in identifying ADHD with 91% accuracy, 94% precision, and 99% specificity. Conclusion: Using Streamlit and Python a web application was developed based on the findings of the analysis. The application will assist parents/guardians and school officials in detecting mental illnesses early in their children and adolescents using signs and symptoms to start the treatment at the earliest convenience.