Independent Component Analysis Model Research Articles

A conditional generative adversarial networks and Yolov5 Darknet -based skin lesion localization and classification using independent component analysis model

Automated diagnosis systems significantly improve the early detection of skin cancer, thereby contributing to its successful treatment. Skin lesion segmentation represents a crucial yet challenging issue within the realm of computer-assisted dermoscopy image analysis. Numerous segmentation techniques reliant on convolutional neural networks falter in achieving precise lesion boundaries due to the diminishing spatial scale of feature maps as they traverse through network layers. Nonetheless, neural networks demand a substantial amount of data, which presents a notable obstacle in the medical field. More data, and possibly a pre-trained model tailored to medical images, are imperative. Researchers have resorted to extensive augmentation and preprocessing methods, as well as fine-tuning the network using pre-trained models on unrelated images, in order to address the challenge of lesion segmentation as the initial step in skin cancer analysis. This arises from the fact that datasets are limited in size and encompass a diverse array of images, varying in terms of lighting, color, scale, and markings. The proposed method addresses the issue of precisely locating and classifying skin lesions in dermatological images. The architecture of Conditional Generative Adversarial Networks (cGAN) enables the generation of high-resolution, contextually pertinent images, thereby enhancing the discriminative properties of skin lesions. YOLOv5 Darknet has an effective object detection mechanism that aids in precise lesion localization. Independent component analysis (ICA), a signal processing technique that identifies statistically separate components from diverse data sources, is employed to improve the model's robustness and interpretability further. This facilitates extracting latent features from the combined dermatological images and data collection, thereby enhancing the model's understanding of complex lesion characteristics. According to experimental results on an extensive dermatological data set, the proposed cGAN-YOLOv5 technique is effective. In pinpointing and classifying lesion locations, the hybrid model outperforms the competition. The synergistic integration of cGANs, YOLOv5 Darknet, and ICA demonstrates the effectiveness of multifaceted integration in enhancing medical image analysis. Using high-resolution feature blocks and a noise-suppressing attention method, cGAN-YOLOv5 is proposed. Therefore, the proposed strategy segments skin lesions effectively.

An integrated approach combining randomized kernel PCA, Gaussian mixture modeling and ICA for fault detection in non-linear processes

Principal component analysis (PCA) and independent component analysis (ICA), as well as their kernel extensions, have been widely applied in the past for industrial fault detection with Gaussian or non-Gaussian process data with linear or non-linear characteristics. Kernel-based techniques lead to computational complexity due to the high dimensionality of the dataset in the feature space. In this work, a randomization approach is used to obtain a low-rank approximation of the high-dimensional kernel matrix. A hybrid machine learning technique is proposed that integrates randomized kernel PCA (RKPCA) with ICA and Gaussian mixture modeling (GMM). The proposed approach, ICA-RKPCA-GMM, addresses the Gaussian and non-Gaussian characteristics of non-linear process data. Another hybrid algorithm combining three basic techniques of ICA, PCA and GMM is also developed (ICA-PCA-GMM). The fault detection performances of the proposed techniques (ICA-RKPCA-GMM and ICA-PCA-GMM) are compared with PCA, ICA, KPCA and combined ICA-PCA techniques by applying the techniques to two benchmark systems. Monitoring performances were evaluated by determining the false alarm rate and fault detection rate for different types of process and sensor faults. The simulation results show that the proposed ICA-RKPCA-GMM approach yields better results than individual ICA, PCA and KPCA techniques, the combined ICA-PCA and the proposed ICA-PCA-GMM technique.

Novel approach for ECG separation using adaptive constrained IVABMGGMM

In this paper, we introduce the constrained independent vector analysis integrated with the bounded multivariate generalized Gaussian mixture model (cIVABMGGMM) to tackle the limitations of independent component analysis (ICA) when applied to multivariate data, accompanied by its adaptive version, the acIVABMGGMM, aimed at alleviating associated constraints. The acIVABMGGMM employs a full covariance matrix that considers feature correlations, effectively addressing the challenges posed by ICA and independent vector analysis (IVA) models when analyzing multivariate data. The innovative acIVABMGGMM framework merges the adaptability inherent in data-driven methods with the capability to manage noise and other artifacts often encountered in model-based approaches. This technique effectively employs prior knowledge to guide the solution, avoiding the imposition of inaccurate constraints. To overcome these challenges, our set of two constrained methods incorporates prior source information into the IVA model, effectively mitigating its limitations in data with a high number of sources. We assess the efficacy of our proposed models through three distinct ECG separation experiments, which include heartbeat separation, fetal ECG extraction, and arrhythmia detection. Notably, the performance metrics demonstrate the superiority of our models over the baseline approaches in the conducted experiments.

Application of independent component analysis-based dimensionality reduction technique in effective information extraction of high-dimensional high-frequency data

Abstract In this paper, after analyzing the existing independent component analysis algorithms, the high-dimensional high-frequency data are preprocessed by whitening, and the objective functions of crag, negative entropy, approximate negative entropy, and mutual information are set. The independent component analysis model is designed to separate the independence between signals by maximizing non-Gaussianity, estimating great likelihood, and minimizing mutual information measures. In order to verify that the dimensionality reduction technique based on independent component analysis can effectively extract high-dimensional, high-frequency data information, it is applied to the industry’s closing index data and financial market data. The results show that the stock price synthesized using the six major independent factors is almost the same as the original stock price trend of DG, and the difference in stock price evaluation is within 5, indicating that the six major independent factors play a decisive role in the stock price trend. The study shows that the dimensionality reduction technique based on independent component analysis can analyze the volatility of stock prices and obtain more effective information from high-dimensional, high-frequency data.

Characterizing Typhoon-related Posttraumatic Stress Disorder Based on Multimodal Fusion of Structural, Diffusion, and Functional Magnetic Resonance Imaging

Diagnosing posttraumatic stress disorder (PTSD) using only single-modality images is controversial. We aimed to use multimodal magnetic resonance imaging (MRI) combining structural, diffusion, and functional MRI to possibly provide a more comprehensive viewpoint on the decisive characteristics of PTSD patients. Typhoon-exposed individuals with (n = 26) and without PTSD (n = 32) and healthy volunteers (n = 30) were enrolled. Five MRI features from three modalities, including two resting-state functional MRI (rs-fMRI) features (amplitude of low-frequency fluctuation, ALFF; and regional homogeneity, ReHo), one structural MRI feature (gray matter density, GM), and two diffusion tensor imaging (DTI) features (fractional anisotropy, FA; and mean diffusivity, MD) were investigated simultaneously with a multimodal canonical correlation analysis + joint independent component analysis model to identify abnormalities in the PTSD brain. We identified statistical differences between PTSD patients and healthy controls in terms of 1 rs-fMRI (ALFF, ReHo) alterations in the superior frontal gyrus, precuneus, inferior parietal lobule (IPL), anterior cingulate cortex (ACC), and posterior cingulate cortex (PCC), 2 DTI (FA, MD) changes in the pons, genu, and splenium of the corpus callosum, and 3 Structural MRI abnormalities in the precuneus, IPL, ACC, and PCC. A novel ReHo component was found to distinguish PTSD and trauma-exposed controls, including the precuneus, IPL, middle frontal gyrus, middle occipital gyrus, and cerebellum. This study reveals that PTSD individuals exhibit intertwined functional and structural anomalies within the default mode network. Some alterations within this network may serve as a potential marker to distinguish between PTSD patients and trauma-exposed controls.

Removal of site effects and enhancement of signal using dual projection independent component analysis for pooling multi-site MRI data.

Combining magnetic resonance imaging (MRI) data from multi-site studies is a popular approach for constructing larger datasets to greatly enhance the reliability and reproducibility of neuroscience research. However, the scanner/site variability is a significant confound that complicates the interpretation of the results, so effective and complete removal of the scanner/site variability is necessary to realise the full advantages of pooling multi-site datasets. Independent component analysis (ICA) and general linear model (GLM) based harmonisation methods are the two primary methods used to eliminate scanner/site effects. Unfortunately, there are challenges with both ICA-based and GLM-based harmonisation methods to remove site effects completely when the signals of interest and scanner/site effects-related variables are correlated, which may occur in neuroscience studies. In this study, we propose an effective and powerful harmonisation strategy that implements dual projection (DP) theory based on ICA to remove the scanner/site effects more completely. This method can separate the signal effects correlated with site variables from the identified site effects for removal without losing signals of interest. Both simulations and vivo structural MRI datasets, including a dataset from Autism Brain Imaging Data Exchange II and a travelling subject dataset from the Strategic Research Program for Brain Sciences, were used to test the performance of a DP-based ICA harmonisation method. Results show that DP-based ICA harmonisation has superior performance for removing site effects and enhancing the sensitivity to detect signals of interest as compared with GLM-based and conventional ICA harmonisation methods.

Enhancing Wind Turbine Performance: Statistical Detection of Sensor Faults Based on Improved Dynamic Independent Component Analysis

Efficient detection of sensor faults in wind turbines is essential to ensure the reliable operation and performance of these renewable energy systems. This paper presents a novel semi-supervised data-based monitoring technique for fault detection in wind turbines using SCADA (supervisory control and data acquisition) data. Unlike supervised methods, the proposed approach does not require labeled data, making it cost-effective and practical for wind turbine monitoring. The technique builds upon the Independent Component Analysis (ICA) approach, effectively capturing non-Gaussian features. Specifically, a dynamic ICA (DICA) model is employed to account for the temporal dynamics and dependencies in the observed signals affected by sensor faults. The fault detection process integrates fault indicators based on I2d, I2e, and squared prediction error (SPE), enabling the identification of different types of sensor faults. The fault indicators are combined with a Double Exponential Weighted Moving Average (DEWMA) chart, known for its superior performance in detecting faults with small magnitudes. Additionally, the approach incorporates kernel density estimation to establish nonparametric thresholds, increasing flexibility and adaptability to different data types. This study considers various types of sensor faults, including bias sensor faults, precision degradation faults, and freezing sensor faults, for evaluation. The results demonstrate that the proposed approach outperforms PCA and traditional ICA-based methods. It achieves a high detection rate, accurately identifying faults while reducing false alarms. It could be a promising technique for proactive maintenance, optimizing the performance and reliability of wind turbine systems.

An independent component analysis based correlation coefficient method for internal short-circuit fault diagnosis of battery-powered intelligent transportation systems

Battery-powered intelligent transportation systems (BPITS) are growing at an alarming rate, and a problem that cannot be ignored is that the safety of battery packs is currently the weak link in the security of BPITS. However, developing a fast and reliable diagnostic method is not easy due to the large number of battery cells and the difficulty to guarantee consistency among the individual cells. To better resolve this problem, a new fault diagnosis scheme based on voltage correlation coefficient (VCC) and independent component analysis (ICA) model is proposed in this paper. Where VCC is used to extract the fault signature of the battery pack independent of the cell inconsistency; and ICA is used to model the non-Gaussian distributed high-dimensional VCC signals in parallel and derive a statistical-based fault detection index. The kernel density estimation-based algorithm is used to calculate the theoretical threshold and to determine the optimal window width for VCC. In particular, a variable reconstruction-based fault isolation algorithm is proposed for accurate and reliable determination of the location of the faulty cell to prevent the smearing effect. Experimental results conducted on a real battery testbed under different conditions and configurations show that the new method provides significant improvements in both detection latency and positioning reliability in diagnosing internal short-circuit (ISC) faults with different failure grades.

Simultaneous fault detection and isolation based on multi-task long short-term memory neural networks

Fault detection and isolation are two significant problems in process monitoring. However, due to the complicated relationships between the process variables, it becomes a challenging problem to build a model to capture the complicated relationship between process variables and perform fault detection and isolation based on the model. This paper proposes an advanced model, called multi-task long short-term memory (MTLSTM) neural networks, that can capture the most complicated relationships between process variables and a simultaneous fault detection and isolation approach based on MTLSTM neural networks. The proposed simultaneous fault detection and isolation approach takes a sparse autoencoder (SAE) to extract features from the measurements of process variables. Then, a long short-term memory neural network is trained on the extracted features with a multi-task learning strategy. Compared with the traditional multivariate statistical models (MVSM) such as principal component analysis and independent component analysis models, the MTLSTM neural networks can capture the nonlinear autocorrelations of process variables and correlations between process variables and perform fault detection and isolation for an industrial process simultaneously. To demonstrate the advantages and superiority of the proposed simultaneous fault detection and isolation approach, a case study on fault detection and isolation for the penicillin fermentation process is carried out. Case study results show that the proposed approach significantly outperforms existing fault detection and isolation approaches.

Unsupervised representation learning of spontaneous MEG data with nonlinear ICA

Resting-state magnetoencephalography (MEG) data show complex but structured spatiotemporal patterns. However, the neurophysiological basis of these signal patterns is not fully known and the underlying signal sources are mixed in MEG measurements. Here, we developed a method based on the nonlinear independent component analysis (ICA), a generative model trainable with unsupervised learning, to learn representations from resting-state MEG data. After being trained with a large dataset from the Cam-CAN repository, the model has learned to represent and generate patterns of spontaneous cortical activity using latent nonlinear components, which reflects principal cortical patterns with specific spectral modes. When applied to the downstream classification task of audio-visual MEG, the nonlinear ICA model achieves competitive performance with deep neural networks despite limited access to labels. We further validate the generalizability of the model across different datasets by applying it to an independent neurofeedback dataset for decoding the subject's attentional states, providing a real-time feature extraction and decoding mindfulness and thought-inducing tasks with an accuracy of around 70% at the individual level, which is much higher than obtained by linear ICA or other baseline methods. Our results demonstrate that nonlinear ICA is a valuable addition to existing tools, particularly suited for unsupervised representation learning of spontaneous MEG activity which can then be applied to specific goals or tasks when labelled data are scarce.

Sparse Bayesian modeling of hierarchical independent component analysis: Reliable estimation of individual differences in brain networks.

Independent component analysis (ICA) is one of the leading approaches for studying brain functional networks. There is increasing interest in neuroscience studies to investigate individual differences in brain networks and their association with demographic characteristics and clinical outcomes. In this work, we develop a sparse Bayesian group hierarchical ICA model that offers significant improvements over existing ICA techniques for identifying covariate effects on the brain network. Specifically, we model the population-level ICA source signals for brain networks using a Dirichlet process mixture. To reliably capture individual differences on brain networks, we propose sparse estimation of the covariate effects in the hierarchical ICA model via a horseshoe prior. Through extensive simulation studies, we show that our approach performs considerably better in detecting covariate effects in comparison with the leading group ICA methods. We then perform an ICA decomposition of a between-subject meditation study. Our method is able to identify significant effects related to meditative practice in brain regions that are consistent with previous research into the default mode network, whereas other group ICA approaches find few to noeffects.

Noise subtraction from KAGRA O3GK data using Independent Component Analysis

During April 7–21 2020, KAGRA conducted its first scientific observation in conjunction with the GEO600 detector. The dominant noise sources during this run were found to be suspension control noise in the low-frequency range and acoustic noise in the mid-frequency range. In this study, we show that their contributions in the observational data can be reduced by a signal processing method called independent component analysis (ICA). The model of ICA is extended from that studied in the initial KAGRA data analysis to account for frequency dependence, while the linearity and stationarity of the coupling between the interferometer and the noise sources are still assumed. We identify optimal witness sensors in the application of ICA, leading to successful mitigation of these two dominant contributions. We also analyze the stability of the transfer functions for the entire two weeks of data to investigate the applicability of the proposed subtraction method in gravitational wave searches.

Multi-model order spatially constrained ICA reveals highly replicable group differences and consistent predictive results from resting data: A large N fMRI schizophrenia study

Brain functional networks identified from resting functional magnetic resonance imaging (fMRI) data have the potential to reveal biomarkers for brain disorders, but studies of complex mental illnesses such as schizophrenia (SZ) often yield mixed results across replication studies. This is likely due in part to the complexity of the disorder, the short data acquisition time, and the limited ability of the approaches for brain imaging data mining. Therefore, the use of analytic approaches which can both capture individual variability while offering comparability across analyses is highly preferred. Fully blind data-driven approaches such as independent component analysis (ICA) are hard to compare across studies, and approaches that use fixed atlas-based regions can have limited sensitivity to individual sensitivity. By contrast, spatially constrained ICA (scICA) provides a hybrid, fully automated solution that can incorporate spatial network priors while also adapting to new subjects. However, scICA has thus far only been used with a single spatial scale (ICA dimensionality, i.e., ICA model order). In this work, we present an approach using multi-objective optimization scICA with reference algorithm (MOO-ICAR) to extract subject-specific intrinsic connectivity networks (ICNs) from fMRI data at multiple spatial scales, which also enables us to study interactions across spatial scales. We evaluate this approach using a large N (N > 1,600) study of schizophrenia divided into separate validation and replication sets. A multi-scale ICN template was estimated and labeled, then used as input into scICA which was computed on an individual subject level. We then performed a subsequent analysis of multiscale functional network connectivity (msFNC) to evaluate the patient data, including group differences and classification. Results showed highly consistent group differences in msFNC in regions including cerebellum, thalamus, and motor/auditory networks. Importantly, multiple msFNC pairs linking different spatial scales were implicated. The classification model built on the msFNC features obtained up to 85% F1 score, 83% precision, and 88% recall, indicating the strength of the proposed framework in detecting group differences between schizophrenia and the control group. Finally, we evaluated the relationship of the identified patterns to positive symptoms and found consistent results across datasets. The results verified the robustness of our framework in evaluating brain functional connectivity of schizophrenia at multiple spatial scales, implicated consistent and replicable brain networks, and highlighted a promising approach for leveraging resting fMRI data for brain biomarker development.

Big Medical Image Analysis: Alzheimer’s Disease Classification Using Convolutional Autoencoder

Deep learning-based analysis is a noticeable topic in recent years. The enormous success of deep learning is now combined with big data analytics to provide an open platform to the health care industry for a better diagnosis of any disease. In this paper, we described the convolutional autoencoder technique that reduces the complexity of radiologists through a brief study of Alzheimer's MRI data which led to a rise in data-driven medical research for a better diagnosis. In this research, we have compared the effects of two techniques: convolutional autoencoder (CANN) and independent component analysis (ICA), and discovered that CANN has a higher accuracy of 98.8% and outperforms ICA models in terms of convergence speed.

Fault detection in multimodal processes based on the local entropy double subspace

Recently, principal component analysis (PCA) and independent component analysis (ICA) have attracted increasing attention for fault detection in industrial processes. However, industrial processes usually have multimodal characteristics, in which case, the performance of traditional PCA and ICA methods is poor. A fault detection method based on the local entropy double subspace (LEDS) is proposed that effectively supervises the multimodal process with multiple steady modes. The proposed method uses the local probability density (LPD) to eliminate the multimodal characteristics of the data. Furthermore, local entropy (LE) is applied to mine the statistical information associated with the process data. The process variables are classified into Gaussian and non-Gaussian subspaces by the Kolmogorov–Smirnov (KS) test. PCA and ICA models are built in Gaussian and non-Gaussian subspaces, respectively, for fault detection. A Bayesian decision method is employed to transform the detection results into fault probabilities, and the detection results are combined with the final statistical information. This method effectively eliminates the multicentre structure of the data and improves the fault detection performance of traditional PCA and ICA. The proposed method is applied to the Tennessee Eastman multimodal process. The simulation results show that the fault detection rate of LEDS is better than the rates of traditional methods when the false alarm rate is low for most faults.

An accurate fake news detection approach based on a Levy flight honey badger optimized convolutional neural network model

SummaryPeople can quickly acquire the news through a variety of sources, including websites, blogs, and social media, among others. The spread of fake news has become easier as a result of the availability of these platforms. Anybody with access to these networks generates and distributes fake news for professional or personal gain. Numerous studies relying on supervised and unsupervised learning techniques are available to address the issue of recognizing fake news. All of those studies, though, have one flaw: they all deliver mostly inaccurate or unmatched results. Poor accuracy is attributed to a variety of factors, including imbalanced datasets, inefficient parameter tuning, poor feature selection, and so on. To tackle these issues, we proposed a novel approach for fake news detection. Initially, the data were obtained from the ISOT dataset and data cleaning is performed. After that, preprocessing is done which includes three major steps such as stemming, stop word removal, and tokenization are carried out. Next to preprocessing, various features that involve name entity recognition‐based features are selected during feature extraction. From this, the short dimensionality features are selected with the help of the ensemble modified independent component analysis model. Finally, the hybrid convolutional neural network‐based Levy flight‐based honey badger algorithm detects fake news. The experiments are simulated using python software with various performance metrics such as accuracy, specificity, sensitivity, precision, and F‐scores to validate the performance of the proposed method. The proposed model offers a precision, recall, and accuracy value of 95%, 97%, and 98% when evaluated with the ISOT dataset. When compared to the existing state‐of‐art methods, the proposed method yielded superior detection results and higher accuracy rates.

An efficient and simple SERS approach for trace analysis of tetrahydrocannabinol and cannabinol and multi-cannabinoid detection

Many countries have legalized cannabis and its derived products for multiple purposes. Consequently, it has become necessary to develop a rapid, effective, and reliable tool for detecting delta-9-tetrahydrocannabinol (THC) and cannabinol (CBN), which are important biologically active compounds in cannabis. Herein, we have fabricated SERS chips by using glancing angle deposition and tuned dimensions of silver nanorods (AgNRs) for detecting THC and CBN at low concentrations. Experimental and computational results showed that the AgNR substrate with film thickness (or nanorod length) of 150 nm, corresponding to nanorod diameter of 79 nm and gap between nanorods of 23 nm, can effectively sense trace THC and CBN with good reproducibility and sensitivity. Due to limited spectral studies of the cannabinoids in previous reports, this work also explored towards identifying characteristic Raman lines of THC and CBN. This information is critical to further reliable data analysis and interpretation. Moreover, multianalyte detection of THC and CBN in a mixture was successfully demonstrated by applying an open-source independent component analysis (ICA) model. The overall method is fast, sensitive, and reliable for sensing trace THC and CBN. The SERS chip-based method and spectral results here are useful for a variety of cannabis testing applications, such as product screening and forensic investigation.

Compressive independent component analysis: theory and algorithms

Abstract Compressive learning forms the exciting intersection between compressed sensing and statistical learning where one exploits sparsity of the learning model to reduce the memory and/or computational complexity of the algorithms used to solve the learning task. In this paper, we look at the independent component analysis (ICA) model through the compressive learning lens. In particular, we show that solutions to the cumulant-based ICA model have a particular structure that induces a low-dimensional model set that resides in the cumulant tensor space. By showing that a restricted isometry property holds for random cumulants e.g. Gaussian ensembles, we prove the existence of a compressive ICA scheme. Thereafter, we propose two algorithms of the form of an iterative projection gradient and an alternating steepest descent algorithm for compressive ICA, where the order of compression asserted from the restricted isometry property is realized through empirical results. We provide analysis of the CICA algorithms including the effects of finite samples. The effects of compression are characterized by a trade-off between the sketch size and the statistical efficiency of the ICA estimates. By considering synthetic and real datasets, we show the substantial memory gains achieved over well-known ICA algorithms by using one of the proposed CICA algorithms.

Direct ICA on data tensor via random matrix modeling

Independent Component Analysis (ICA) is a fundamental method for Blind Source Separation (BSS). Classical ICA takes data matrix input formed by vector data. This paper focuses on ICA for BSS with third-order data tensor input formed by matrix data, such as 2D images. Two approaches exist for this problem. The first approach reshapes each matrix into a vector to apply classical ICA, with structural information lost. The second approach unfolds a data tensor into a data matrix along different modes to perform classical ICA mode-wise, which partially preserves structures but has strong or ill BSS assumptions. This paper proposes a third approach via RAndom Matrix ICA (RAMICA) modeling. RAMICA works on data tensor directly, without vectorization or unfolding, and preserves row or column structures under more general BSS assumptions. We develop the RAMICA model, algorithm, and related theories via defining new statistics for random matrices and new procedures for whitening and independent component estimation. We study the identifiability, higher-order extension, and relationships with existing methods. Experiments on both synthetic and real data show superior BSS performance of RAMICA over competing methods and offer insights on the trade-offs between different factors.

Prediction of Recovery from Traumatic Brain Injury with EEG Power Spectrum in Combination of Independent Component Analysis and RUSBoost Model

The computational electroencephalogram (EEG) is recently garnering significant attention in examining whether the quantitative EEG (qEEG) features can be used as new predictors for the prediction of recovery in moderate traumatic brain injury (TBI). However, the brain’s recorded electrical activity has always been contaminated with artifacts, which in turn further impede the subsequent processing steps. As a result, it is crucial to devise a strategy for meticulously flagging and extracting clean EEG data to retrieve high-quality discriminative features for successful model development. This work proposed the use of multiple artifact rejection algorithms (MARA), which is an independent component analysis (ICA)-based algorithm, to eliminate artifacts automatically, and explored their effects on the predictive performance of the random undersampling boosting (RUSBoost) model. Continuous EEG were acquired using 64 electrodes from 27 moderate TBI patients at four weeks to one-year post-accident. The MARA incorporates an artifact removal stage based on ICA prior to RUSBoost, SVM, DT, and k-NN classification. The area under the curve (AUC) of RUSBoost was higher in absolute power spectral density (PSD) in AUCδ = 0.75, AUC α = 0.73 and AUCθ = 0.71 bands than SVM, DT, and k-NN. The MARA has provided a good generalization performance of the RUSBoost prediction model.