Search Results (17)

Solar energy is a critical renewable energy source, with solar arrays or photovoltaic systems widely used to convert solar energy into electrical energy. However, solar array systems can develop faults and may exhibit poor performance. Diagnosing and resolving faults within these systems promptly is crucial to ensure reliability and efficiency in energy generation. Autoencoders and their variants have gained popularity in recent studies for detecting and diagnosing faults in solar arrays. However, traditional autoencoder models often struggle to capture the spatial and temporal relationships present in photovoltaic sensor data. This paper introduces a deep learning model that combines a graph convolutional network with a variational autoencoder to diagnose faults in solar arrays. The graph convolutional network effectively learns from spatial and temporal sensor data, significantly improving fault detection performance. We evaluated the proposed deep learning model on a recently published solar array dataset for an integrated power probability table mode. The experimental results show that the model achieves a fault detection rate exceeding 95% and outperforms the conventional autoencoder models. We also identified faulty components by analyzing the model’s reconstruction error for each feature, and we validated the analysis through the Kolmogorov–Smirnov test and noise injection techniques. Full article

Disease recognition has been revolutionized by autonomous systems in the rapidly developing field of medical technology. A crucial aspect of diagnosis involves the visual assessment and enumeration of white blood cells in microscopic peripheral blood smears. This practice yields invaluable insights into a patient’s health, enabling the identification of conditions of blood malignancies such as leukemia. Early identification of leukemia subtypes is paramount for tailoring appropriate therapeutic interventions and enhancing patient survival rates. However, traditional diagnostic techniques, which depend on visual assessment, are arbitrary, laborious, and prone to errors. The advent of ML technologies offers a promising avenue for more accurate and efficient leukemia classification. In this study, we introduced a novel approach to leukemia classification by integrating advanced image processing, diverse dataset utilization, and sophisticated feature extraction techniques, coupled with the development of TL models. Focused on improving accuracy of previous studies, our approach utilized Kaggle datasets for binary and multiclass classifications. Extensive image processing involved a novel LoGMH method, complemented by diverse augmentation techniques. Feature extraction employed DCNN, with subsequent utilization of extracted features to train various ML and TL models. Rigorous evaluation using traditional metrics revealed Inception-ResNet’s superior performance, surpassing other models with F1 scores of 96.07% and 95.89% for binary and multiclass classification, respectively. Our results notably surpass previous research, particularly in cases involving a higher number of classes. These findings promise to influence clinical decision support systems, guide future research, and potentially revolutionize cancer diagnostics beyond leukemia, impacting broader medical imaging and oncology domains. Full article

This study focuses on leveraging data-driven techniques to diagnose brain tumors through magnetic resonance imaging (MRI) images. Utilizing the rule of deep learning (DL), we introduce and fine-tune two robust frameworks, ResNet 50 and Inception V3, specifically designed for the classification of brain MRI images. Building upon the previous success of ResNet 50 and Inception V3 in classifying other medical imaging datasets, our investigation encompasses datasets with distinct characteristics, including one with four classes and another with two. The primary contribution of our research lies in the meticulous curation of these paired datasets. We have also integrated essential techniques, including Early Stopping and ReduceLROnPlateau, to refine the model through hyperparameter optimization. This involved adding extra layers, experimenting with various loss functions and learning rates, and incorporating dropout layers and regularization to ensure model convergence in predictions. Furthermore, strategic enhancements, such as customized pooling and regularization layers, have significantly elevated the accuracy of our models, resulting in remarkable classification accuracy. Notably, the pairing of ResNet 50 with the Nadam optimizer yields extraordinary accuracy rates, reaching 99.34% for gliomas, 93.52% for meningiomas, 98.68% for non-tumorous images, and 97.70% for pituitary tumors. These results underscore the transformative potential of our custom-made approach, achieving an aggregate testing accuracy of 97.68% for these four distinct classes. In a two-class dataset, Resnet 50 with the Adam optimizer excels, demonstrating better precision, recall, F1 score, and an overall accuracy of 99.84%. Moreover, it attains perfect per-class accuracy of 99.62% for ‘Tumor Positive’ and 100% for ‘Tumor Negative’, underscoring a remarkable advancement in the realm of brain tumor categorization. This research underscores the innovative possibilities of DL models and our specialized optimization methods in the domain of diagnosing brain cancer from MRI images. Full article

Gait recognition, also known as walking pattern recognition, has expressed deep interest in the computer vision and biometrics community due to its potential to identify individuals from a distance. It has attracted increasing attention due to its potential applications and non-invasive nature. Since 2014, deep learning approaches have shown promising results in gait recognition by automatically extracting features. However, recognizing gait accurately is challenging due to the covariate factors, complexity and variability of environments, and human body representations. This paper provides a comprehensive overview of the advancements made in this field along with the challenges and limitations associated with deep learning methods. For that, it initially examines the various gait datasets used in the literature review and analyzes the performance of state-of-the-art techniques. After that, a taxonomy of deep learning methods is presented to characterize and organize the research landscape in this field. Furthermore, the taxonomy highlights the basic limitations of deep learning methods in the context of gait recognition. The paper is concluded by focusing on the present challenges and suggesting several research directions to improve the performance of gait recognition in the future. Full article

Due to its high transmissibility, the COVID-19 pandemic has placed an unprecedented burden on healthcare systems worldwide. X-ray imaging of the chest has emerged as a valuable and cost-effective tool for detecting and diagnosing COVID-19 patients. In this study, we developed a deep learning model using transfer learning with optimized DenseNet-169 and DenseNet-201 models for three-class classification, utilizing the Nadam optimizer. We modified the traditional DenseNet architecture and tuned the hyperparameters to improve the model’s performance. The model was evaluated on a novel dataset of 3312 X-ray images from publicly available datasets, using metrics such as accuracy, recall, precision, F1-score, and the area under the receiver operating characteristics curve. Our results showed impressive detection rate accuracy and recall for COVID-19 patients, with 95.98% and 96% achieved using DenseNet-169 and 96.18% and 99% using DenseNet-201. Unique layer configurations and the Nadam optimization algorithm enabled our deep learning model to achieve high rates of accuracy not only for detecting COVID-19 patients but also for identifying normal and pneumonia-affected patients. The model’s ability to detect lung problems early on, as well as its low false-positive and false-negative rates, suggest that it has the potential to serve as a reliable diagnostic tool for a variety of lung diseases. Full article

With the proliferation of the internet, social networking sites have become a primary source of user-generated content, including vast amounts of information about medications, diagnoses, treatments, and disorders. Comments on previously used medicines, contained within these data, can be leveraged to identify crucial adverse drug reactions, and machine learning (ML) approaches such as sentiment analysis (SA) can be employed to derive valuable insights. However, given the sheer volume of comments, it is often impractical for consumers to manually review all of them before determining a purchase decision. Therefore, drug assessments can serve as a valuable source of medical information for both healthcare professionals and the general public, aiding in decision making and improving public monitoring systems by revealing collective experiences. Nonetheless, the unstructured and linguistic nature of the comments poses a significant challenge for effective categorization, with previous studies having utilized machine and deep learning (DL) algorithms to address this challenge. Despite both approaches showing promising results, DL classifiers outperformed ML classifiers in previous studies. Therefore, the objective of our study was to improve upon earlier research by applying SA to medication reviews and training five ML algorithms on two distinct feature extractions and four DL classifiers on two different word-embedding approaches to obtain higher categorization scores. Our findings indicated that the random forest trained on the count vectorizer outperformed all other ML algorithms, achieving an accuracy and F1 score of 96.65% and 96.42%, respectively. Furthermore, the bidirectional LSTM (Bi-LSTM) model trained on GloVe embedding resulted in an even better accuracy and F1 score, reaching 97.40% and 97.42%, respectively. Hence, by utilizing appropriate natural language processing and ML algorithms, we were able to achieve superior results compared to earlier studies. Full article

This paper proposes a new deep learning (DL) framework for the analysis of lung diseases, including COVID-19 and pneumonia, from chest CT scans and X-ray (CXR) images. This framework is termed optimized DenseNet201 for lung diseases (LDDNet). The proposed LDDNet was developed using additional layers of 2D global average pooling, dense and dropout layers, and batch normalization to the base DenseNet201 model. There are 1024 Relu-activated dense layers and 256 dense layers using the sigmoid activation method. The hyper-parameters of the model, including the learning rate, batch size, epochs, and dropout rate, were tuned for the model. Next, three datasets of lung diseases were formed from separate open-access sources. One was a CT scan dataset containing 1043 images. Two X-ray datasets comprising images of COVID-19-affected lungs, pneumonia-affected lungs, and healthy lungs exist, with one being an imbalanced dataset with 5935 images and the other being a balanced dataset with 5002 images. The performance of each model was analyzed using the Adam, Nadam, and SGD optimizers. The best results have been obtained for both the CT scan and CXR datasets using the Nadam optimizer. For the CT scan images, LDDNet showed a COVID-19-positive classification accuracy of 99.36%, a 100% precision recall of 98%, and an F1 score of 99%. For the X-ray dataset of 5935 images, LDDNet provides a 99.55% accuracy, 73% recall, 100% precision, and 85% F1 score using the Nadam optimizer in detecting COVID-19-affected patients. For the balanced X-ray dataset, LDDNet provides a 97.07% classification accuracy. For a given set of parameters, the performance results of LDDNet are better than the existing algorithms of ResNet152V2 and XceptionNet. Full article

An automated sleep stage categorization can readily face noise-contaminated EEG recordings, just as other signal processing applications. Therefore, the denoising of the contaminated signals is inevitable to ensure a reliable analysis of the EEG signals. In this research work, an empirical mode decomposition is used in combination with stacked autoencoders to conduct automatic sleep stage classification with reliable analytical performance. Due to the decomposition of the composite signal into several intrinsic mode functions, empirical mode decomposition offers an effective solution for denoising non-stationary signals such as EEG. Preliminary results showed that through these intrinsic modes, a signal with a high signal-to-noise ratio can be obtained, which can be used for further analysis with confidence. Therefore, later, when statistical features were extracted from the denoised signals and were classified using stacked autoencoders, improved results were obtained for Stage 1, Stage 2, Stage 3, Stage 4, and REM stage EEG signals using this combination. Full article

Statistical features extraction from bearing fault signals requires a substantial level of knowledge and domain expertise. Furthermore, existing feature extraction techniques are mostly confined to selective feature extraction methods namely, time-domain, frequency-domain, or time-frequency domain statistical parameters. Vibration signals of bearing fault are highly non-linear and non-stationary making it cumbersome to extract relevant information for existing methodologies. This process even became more complicated when the bearing operates at variable speeds and load conditions. To address these challenges, this study develops an autonomous diagnostic system that combines signal-to-image transformation techniques for multi-domain information with convolutional neural network (CNN)-aided multitask learning (MTL). To address variable operating conditions, a composite color image is created by fusing information from multi-domains, such as the raw time-domain signal, the spectrum of the time-domain signal, and the envelope spectrum of the time-frequency analysis. This 2-D composite image, named multi-domain fusion-based vibration imaging (MDFVI), is highly effective in generating a unique pattern even with variable speeds and loads. Following that, these MDFVI images are fed to the proposed MTL-based CNN architecture to identify faults in variable speed and health conditions concurrently. The proposed method is tested on two benchmark datasets from the bearing experiment. The experimental results suggested that the proposed method outperformed state-of-the-arts in both datasets. Full article

Pipeline leakage remains a challenge in various industries. Acoustic emission (AE) technology has recently shown great potential for leak diagnosis. Many AE features, such as root mean square (RMS), peak value, standard deviation, mean value, and entropy, have been suggested to detect leaks. However, background noise in AE signals makes these features ineffective. The present paper proposes a pipeline leak detection technique based on acoustic emission event (AEE) features and a Kolmogorov–Smirnov (KS) test. The AEE features, namely, peak amplitude, energy, rise-time, decay time, and counts, are inherent properties of AE signals and therefore more suitable for recognizing leak attributes. Surprisingly, the AEE features have received negligible attention. According to the proposed technique, the AEE features are first extracted from the AE signals. For this purpose, a sliding window was used with an adaptive threshold so that the properties of both burst- and continuous-type emissions can be retained. The AEE features form distribution that change its shape when the pipeline condition changes from normal to leakage. The AEE feature distributions for leak and healthy conditions were discriminated using the two-sample KS test, and a pipeline leak indicator (PLI) was obtained. The experimental results demonstrate that the developed PLI accurately distinguishes the leak and no-leak conditions without any prior leak information and it performs better than the traditional features such as mean, variance, RMS, and kurtosis. Full article

In this paper, an explainable AI-based fault diagnosis model for bearings is proposed with five stages, i.e., (1) a data preprocessing method based on the Stockwell Transformation Coefficient (STC) is proposed to analyze the vibration signals for variable speed and load conditions, (2) a statistical feature extraction method is introduced to capture the significance from the invariant pattern of the analyzed data by STC, (3) an explainable feature selection process is proposed by introducing a wrapper-based feature selector—Boruta, (4) a feature filtration method is considered on the top of the feature selector to avoid the multicollinearity problem, and finally, (5) an additive Shapley explanation followed by k-NN is proposed to diagnose and to explain the individual decision of the k-NN classifier for debugging the performance of the diagnosis model. Thus, the idea of explainability is introduced for the first time in the field of bearing fault diagnosis in two steps: (a) incorporating explainability to the feature selection process, and (b) interpretation of the classifier performance with respect to the selected features. The effectiveness of the proposed model is demonstrated on two different datasets obtained from separate bearing testbeds. Lastly, an assessment of several state-of-the-art fault diagnosis algorithms in rotating machinery is included. Full article

Rolling element bearings are a vital part of rotating machines and their sudden failure can result in huge economic losses as well as physical causalities. Popular bearing fault diagnosis techniques include statistical feature analysis of time, frequency, or time-frequency domain data. These engineered features are susceptible to variations under inconsistent machine operation due to the non-stationary, non-linear, and complex nature of the recorded vibration signals. To address these issues, numerous deep learning-based frameworks have been proposed in the literature. However, the logical reasoning behind crack severities and the longer training times needed to identify multiple health characteristics at the same time still pose challenges. Therefore, in this work, a diagnosis framework is proposed that uses higher-order spectral analysis and multitask learning (MTL), while also incorporating transfer learning (TL). The idea is to first preprocess the vibration signals recorded from a bearing to look for distinct patterns for a given fault type under inconsistent working conditions, e.g., variable motor speeds and loads, multiple crack severities, compound faults, and ample noise. Later, these bispectra are provided as an input to the proposed MTL-based convolutional neural network (CNN) to identify the speed and the health conditions, simultaneously. Finally, the TL-based approach is adopted to identify bearing faults in the presence of multiple crack severities. The proposed diagnostic framework is evaluated on several datasets and the experimental results are compared with several state-of-the-art diagnostic techniques to validate the superiority of the proposed model under inconsistent working conditions. Full article

This paper proposes a classification framework for automatic sleep stage detection in both male and female human subjects by analyzing the electroencephalogram (EEG) data of polysomnography (PSG) recorded for three regions of the human brain, i.e., the pre-frontal, central, and occipital lobes. Without considering any artifact removal approach, the residual neural network (ResNet) architecture is used to automatically learn the distinctive features of different sleep stages from the power spectral density (PSD) of the raw EEG data. The residual block of the ResNet learns the intrinsic features of different sleep stages from the EEG data while avoiding the vanishing gradient problem. The proposed approach is validated using the sleep dataset of the Dreams database, which comprises of EEG signals for 20 healthy human subjects, 16 female and 4 male. Our experimental results demonstrate the effectiveness of the ResNet based approach in identifying different sleep stages in both female and male subjects compared to state-of-the-art methods with classification accuracies of 87.8% and 83.7%, respectively. Full article

Feature analysis puts a great impact in determining the various health conditions of mechanical vessels. To achieve balance between traditional feature extraction and the automated feature selection process, a hybrid bag of features (HBoF) is designed for multiclass health state classification of spherical tanks in this paper. The proposed HBoF is composed of (a) the acoustic emission (AE) features and (b) the time and frequency based statistical features. A wrapper-based feature chooser algorithm, Boruta, is utilized to extract the most intrinsic feature set from HBoF. The selective feature matrix is passed to the multi-class k-nearest neighbor (k-NN) algorithm to differentiate among normal condition (NC) and two faulty conditions (FC1 and FC2). Experimental results demonstrate that the proposed methodology generates an average 99.7% accuracy for all working conditions. Moreover, it outperforms the existing state-of-art works by achieving at least 19.4%. Full article

Human stress analysis using electroencephalogram (EEG) signals requires a detailed and domain-specific information pool to develop an effective machine learning model. In this study, a multi-domain hybrid feature pool is designed to identify most of the important information from the signal. The hybrid feature pool contains features from two types of analysis: (a) statistical parametric analysis from the time domain, and (b) wavelet-based bandwidth specific feature analysis from the time-frequency domain. Then, a wrapper-based feature selector, Boruta, is applied for ranking all the relevant features from that feature pool instead of considering only the non-redundant features. Finally, the k-nearest neighbor (k-NN) algorithm is used for final classification. The proposed model yields an overall accuracy of 73.38% for the total considered dataset. To validate the performance of the proposed model and highlight the necessity of designing a hybrid feature pool, the model was compared to non-linear dimensionality reduction techniques, as well as those without feature ranking. Full article