A novel computer-aided diagnosis framework for EEG-based identification of neural diseases

• Published in: Computers in biology and medicine Vol. 138; p. 104922
• Main Authors: Sadiq, Muhammad Tariq, Akbari, Hesam, Siuly, Siuly, Yousaf, Adnan, Rehman, Ateeq Ur
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.11.2021; Elsevier Limited
• Subjects: Accuracy; Adaptability; Algorithms; Classification; Computer-aided diagnosis; Convulsions & seizures; Datasets; Decomposition; Diagnosis; EEG; Electroencephalography; Empirical analysis; Epilepsy; Feature extraction; Geometrical features; Internal Medicine; K-nearest neighbors algorithm; Medical diagnosis; Methods; Morphology; Motivation; Neural diseases; Other; Particle swarm optimization; Rehabilitation; Signal classification; Two dimensional models; Two-dimensional modeling; Wavelet transforms; Neural diseases; Electroencephalography; Two-dimensional modeling; Geometrical features; Computer-aided diagnosis
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2021.104922

Recent advances in electroencephalogram (EEG) signal classification have primarily focused on domain-specific approaches, which impede algorithm cross-discipline capability. This study introduces a new computer-aided diagnosis (CAD) system for the classification of two distinct EEG domains under a unified sequential framework. The key motivation to consider two neural diseases by one framework is to develop a unified algorithm for EEG classification. The main contributions of this study are five-fold. First, EEG signals are decomposed into 10 intrinsic mode functions (IMFs) with the help of empirical wavelet transform. Second, a novel two-dimensional (2D) modeling of IMFs is plotted to visualize the complexity of EEG signals. Third, several new geometrical features are extracted to analyze the dynamic and chaotic essence. Fourth, significant features are selected by binary particle swarm optimization algorithm (B–PSO). Fifth, selected features are fed to the k-nearest neighbor classifier for EEG signal classification purposes. All the experiments are executed on one depression and two epileptic EEG datasets in a leave one out cross-validation strategy. The proposed CAD system provides an average classification accuracy of 93.35% in depression detection, 99.33% for regular against ictal, and 97.33% for interictal versus ictal respectively. The overall empirical analysis authenticates that the proposed CAD outperforms the existing domain-specific methods in terms of classification accuracies and multirole adaptability, thus, can be endorsed as an effective automated neural rehabilitation system. •A Novel two-dimensional modeling is proposed to visualize chaotic behavior of EEG signals.•Novel computationally efficient geometrical features are introduce.•A computerized framework is proposed for automated detection of EEG signals.•The proposed framework is suitable for cross-domain EEG analysis.•Experimental results are better or comparable with state-of-art.