Deep Learning-based Classification Model Research Articles

Identifying sex from pharyngeal images using deep learning algorithm

The pharynx is one of the few areas in the body where blood vessels and immune tissues can readily be observed from outside the body non-invasively. Although prior studies have found that sex could be identified from retinal images using artificial intelligence, it remains unknown as to whether individuals’ sex could also be identified using pharyngeal images. Demographic information and pharyngeal images were collected from patients who visited 64 primary care clinics in Japan for influenza-like symptoms. We trained a deep learning-based classification model to predict reported sex, which incorporated a multiple instance convolutional neural network, on 20,319 images from 51 clinics. Validation was performed using 4869 images from the remaining 13 clinics not used for the training. The performance of the classification model was assessed using the area under the receiver operating characteristic curve. To interpret the model, we proposed a framework that combines a saliency map and organ segmentation map to quantitatively evaluate salient regions. The model achieved the area under the receiver operating characteristic curve of 0.883 (95% CI 0.866–0.900). In subgroup analyses, a substantial improvement in classification performance was observed for individuals aged 20 and older, indicating that sex-specific patterns between women and men may manifest as humans age (e.g., may manifest after puberty). The saliency map suggested the model primarily focused on the posterior pharyngeal wall and the uvula. Our study revealed the potential utility of pharyngeal images by accurately identifying individuals’ reported sex using deep learning algorithm.

A Deep Learning-based Classification Model for Arabic News Tweets Using Bidirectional Long Short-Term Memory Networks

This research develops a classification model for Arabic news tweets using Bidirectional Long Short-Term Memory networks (BiLSTM). Tweets about Arabic news were gathered between August 2016 and August 2020 and divided into five categories. Custom Python scripts, Twitter API and the GetOldTweets3 Python library were used to collect the data. BiLSTM was used to train and test the model. The results indicated an average accuracy, precision, recall, and f1-score of 0.88, 0.92, 0.88, and 0.89, respectively. The results could have practical implications for Arabic machine learning and NLP tasks in research and practice.

A Deep Learning Approach for Fear Recognition on the Edge Based on Two-Dimensional Feature Maps.

Applying affective computing techniques to recognize fear and combining them with portable signal monitors makes it possible to create real-time detection systems that could act as bodyguards when users are in danger. With this aim, this paper presents a fear recognition method based on physiological signals obtained from wearable devices. The procedure involves creating two-dimensional feature maps from the raw signals, using data augmentation and feature selection algorithms, followed by deep learning-based classification models, taking inspiration from those used in image processing. This proposal has been validated with two different datasets, achieving, in WEMAC, WESAD 3-classes, and WESAD 2-classes, F1-score results of 78.13%, 88.07%, and 99.60%, respectively, and 79.90%, 89.12%, and 99.60% in accuracy. Furthermore, the paper demonstrates the feasibility of implementing the proposed method on the Coral Edge TPU device, prepared to make inferences on the edge.

Single-cell classification based on label-free high-resolution optical data of cell adhesion kinetics

Selecting and isolating various cell types is a critical procedure in many applications, including immune therapy, regenerative medicine, and cancer research. Usually, these selection processes involve some labeling or another invasive step potentially affecting cellular functionality or damaging the cell. In the current proof of principle study, we first introduce an optical biosensor-based method capable of classification between healthy and numerous cancerous cell types in a label-free setup. We present high classification accuracy based on the monitored single-cell adhesion kinetic signals. We developed a high-throughput data processing pipeline to build a benchmark database of ~ 4500 single-cell adhesion measurements of a normal preosteoblast (MC3T3-E1) and various cancer (HeLa, LCLC-103H, MDA-MB-231, MCF-7) cell types. Several datasets were used with different cell-type selections to test the performance of deep learning-based classification models, reaching above 70–80% depending on the classification task. Beyond testing these models, we aimed to draw interpretable biological insights from their results; thus, we applied a deep neural network visualization method (grad-CAM) to reveal the basis on which these complex models made their decisions. Our proof-of-concept work demonstrated the success of a deep neural network using merely label-free adhesion kinetic data to classify single mammalian cells into different cell types. We propose our method for label-free single-cell profiling and in vitro cancer research involving adhesion. The employed label-free measurement is noninvasive and does not affect cellular functionality. Therefore, it could also be adapted for applications where the selected cells need further processing, such as immune therapy and regenerative medicine.

A self-supervised learning framework based on masked autoencoder for complex wafer bin map classification

Wafer bin map (WBM) automatic classification is one of the critical challenges for semiconductor intelligent manufacturing. Many deep learning-based classification models have performed well in WBM classification, but all require a large amount of labeled data for training. Since real-world WBMs are highly complex and can be labeled correctly only by seasoned engineers, such requirements undermine the practical value of those methods. Several self-supervised learning methods have recently been proposed for WBM to improve classification performance. However, they still require much labeled data for fine-tuning and are only adapted for binary WBM with a single gross failure area. To address these limitations, this study introduces a self-supervised framework based on masked autoencoder (MAE) for complex WBMs with mixed bin signatures and multiple gross failure area patterns. A patchMC encoder is proposed to improve MAE’s representation ability for complex WBMs with mixed bin signatures. Moreover, the pre-trained MAE encoder with a multi-label classifier fine-tuned by labeled WBMs enables a few-shot classification of complex WBMs with multiple gross failure areas. Experimental validation of the proposed method is performed on a real-world complex WBM dataset from Intel Corporation. The results demonstrate that the proposed method can make good use of unlabeled WBMs and reduce the demand for labeled data to a few-shot level and, at the same time, guarantees a classification accuracy of more than 90%. By comparing MAE with other self-supervised learning methods, MAE outperforms other existing self-supervised methods for WBM data.

Deep learning-based osteochondritis dissecans detection in ultrasound images with humeral capitellum localization.

Osteochondritis dissecans (OCD) of the humeral capitellum is a common cause of elbow disorders, particularly among young throwing athletes. Conservative treatment is the preferred treatment for managing OCD, and early intervention significantly influences the possibility of complete disease resolution. The purpose of this study is to develop a deep learning-based classification model in ultrasound images for computer-aided diagnosis. This paper proposes a deep learning-based OCD classification method in ultrasound images. The proposed method first detects the humeral capitellum detection using YOLO and then estimates the OCD probability of the detected region probability using VGG16. We hypothesis that the performance will be improved by eliminating unnecessary regions. To validate the performance of the proposed method, it was applied to 158 subjects (OCD: 67, Normal: 91) using five-fold-cross-validation. The study demonstrated that the humeral capitellum detection achieved a mean average precision (mAP) of over 0.95, while OCD probability estimation achieved an average accuracy of 0.890, precision of 0.888, recall of 0.927, F1 score of 0.894, and an area under the curve (AUC) of 0.962. On the other hand, when the classification model was constructed for the entire image, accuracy, precision, recall, F1 score, and AUC were 0.806, 0.806, 0.932, 0.843, and 0.928, respectively. The findings suggest the high-performance potential of the proposed model for OCD classification in ultrasonic images. This paper introduces a deep learning-based OCD classification method. The experimental results emphasize the effectiveness of focusing on the humeral capitellum for OCD classification in ultrasound images. Future work should involve evaluating the effectiveness of employing the proposed method by physicians during medical check-ups for OCD.

AI-Assisted Deep NLP-Based Approach for Prediction of Fake News From Social Media Users

Social networking websites are now considered to be the best platforms for the dissemination of news articles. However, information sharing in social media platforms leads to explosion of fake news. Traditional detection methods were focusing on content analysis, while the current researchers examining social features of the news. In this work, we proposed a novel artificial intelligence (AI)-assisted fake news detection with deep natural language processing (NLP) model. The proposed work is characterized in four layers: publisher layer, social media networking layer, enabled edge layer, and cloud layer. In this work, four steps were carried out: 1) data acquisition; 2) information retrieval (IR); 3) NLP-based data processing and feature extraction; and 4) deep learning-based classification model that classifies news articles as fake or real using credibility score of publishers, users, messages, headlines, and so on. Three datasets, such as Buzzface, FakeNewsNet, and Twitter, were used for evaluation of the proposed model, and simulation results were computed. This proposed model obtained an average accuracy of 99.72% and an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$F1$</tex-math> </inline-formula> score of 98.33%, which outperforms other existing methods.

Deep Learning in Smart Healthcare: A GAN-based Approach for Imbalanced Alzheimer's Disease Classification

Alzheimer's disease (AD) is a type of dementia that leads to memory loss and impairment, which afects patients’ lives badly. It is not curable yet, but its progression can be slowed down if detected at earlier stages. In this research study, we propose a transfer learning-based convolutional neural network (CNN) model to classify magnetic resonance imaging (MRI) into one of four stages of Alzheimer's disease. One of the major limitations of the deep learning-based classification model is the non-availability of healthcare datasets related to AD. The widely used Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset has a major class imbalance issue. We propose a generative adversarial network (GAN) based data augmentation technique to overcome the data imbalance. This promotes the investigation of applying GANs to generate synthetic samples for minority classes in Alzheimer's disease datasets to enhance classification performance. The results show the progression in the overall classification process of AD.

An Eye Movement Classification Method based on Cascade Forest.

Eye tracking technology has become increasingly important in scientific research and practical applications. In the field of eye tracking research, analysis of eye movement data is crucial, particularly for classifying raw eye movement data into eye movement events. Current classification methods exhibit considerable variation in adaptability across different participants, and it is necessary to address the issues of class imbalance and data scarcity in eye movement classification. In the current study, we introduce a novel eye movement classification method based on cascade forest (EMCCF), which comprises two modules: (1) a feature extraction module that employs a multi-scale time window method to extract features from raw eye movement data; (2) a classification module that innovatively employs a layered ensemble architecture, integrating the cascade forest structure with ensemble learning principles, specifically for eye movement classification. Consequently, EMCCF not only enhanced the accuracy and efficiency of eye movement classification but also represents an advancement in applying ensemble learning techniques within this domain. Furthermore, experimental results indicated that EMCCF outperformed existing deep learning-based classification models in several metrics and demonstrated robust performance across different datasets and participants.

Design of Barnacle Mating Optimizer with Deep Learning Based Classification Model for Medical X-Ray Images

Design of Barnacle Mating Optimizer with Deep Learning Based Classification Model for Medical X-Ray Images

#319 : Blastocyst Formation Prediction Based on Deep Learning Model from 3-Day Embryo Images in Time-Lapse Incubator Using Data Augmentation

Background and Aims: Current literature suggests that blastocyst ET (Embryo Transfer) at day 5 improves pregnancy outcomes compared with cleavage ET at day 3. However, blastocyst ET poses potential challenges due to the risk of developmental arrest at the cleavage stage. Therefore, accurately predicting blastocyst formation in embryos will help determining optimal days for embryo culture. The aim of our study is to develop a deep learning-based classification model that can predict blastocyst formation based on 3-day embryo images captured from Time-Lapse incubators. Method: A total of 200 embryo images were collected at 72 hours after ICSI (Intracytoplasmic Sperm Injection), with 100 forming blastocysts and 100 failing to do so. The images were annotated by RectLabel for object detection and segmentation, and data augmentation techniques were applied to enhance the dataset. A pre-trained ResNet model was used as the basis for our classifier model, and optimization was performed using the Adam optimizer and BCEWithLogitsLoss as the loss function. Model performance was evaluated through classification accuracy and AUC score. Results: The CNN-based classifier models demonstrated a considerable accuracy of 87.5% and an AUC of 0.896 in predicting blastocyst formation, with sensitivity and specificity of 91.7% and 81.3%, respectively. It was found that without data augmentation, the model degraded to an accuracy of 62.5% and an AUC of 0.631. Conclusion: Our study competently developed a classification model for predicting blastocyst formation using 3-day embryo images from a Time-Lapse incubator and demonstrated the effectiveness of data augmentation techniques in limited data sets. This deep learning-based model may contribute as a beneficial tool in IVF (In Vitro Fertilization) process.

Learning adaptive coarse basis functions of FETI-DP

Domain decomposition methods are successful and highly parallel scalable iterative solution methods for discretized partial differential equations. Nevertheless, for many classes of problems, for example, elliptic partial differential equations with arbitrary coefficient distributions, adaptive coarse spaces are necessary to obtain robustness or, in other words, to guarantee a reliable and fast convergence. Adaptive coarse spaces are usually computed by solving many localized eigenvalue problems related to edges or faces of the domain decomposition. This results in a computationally expensive setup of the domain decomposition preconditioner or system operator. In earlier work, to which the authors have contributed, a deep learning based classification model was used to classify the critical edges or faces where eigenvalue problems have to be solved. In the present paper, we suggest to directly learn the adaptive constraints using a deep feedforward neural network regression model and thus completely skip the computationally most expensive part of the setup, i.e., the solution of local eigenvalue problems. We consider a specific adaptive FETI-DP (Finite Tearing and Interconnecting – Dual Primal) approach and concentrate on stationary diffusion problems in two dimensions with arbitrary coefficient functions with large jumps. As an input for the neural network, we use an image representation of the coefficient function which resolves the structure of the coefficient distribution but is not necessarily identical to the discretization of the partial differential equation. Therefore, our approach is independent of the finite element mesh and can, in principle, be easily extended to other adaptive coarse spaces, problems, and domain decomposition methods. We show the robustness of our method for different problems and the generalization property of our trained neural networks by considering different coefficient distributions not contained in the training set. We also combine the learned constraints with computationally cheap frugal constraints to further improve our approach.

DeepXplainer: An interpretable deep learning based approach for lung cancer detection using explainable artificial intelligence

Background and ObjectiveArtificial intelligence (AI) has several uses in the healthcare industry, some of which include healthcare management, medical forecasting, practical making of decisions, and diagnosis. AI technologies have reached human-like performance, but their use is limited since they are still largely viewed as opaque black boxes. This distrust remains the primary factor for their limited real application, particularly in healthcare. As a result, there is a need for interpretable predictors that provide better predictions and also explain their predictions. MethodsThis study introduces “DeepXplainer”, a new interpretable hybrid deep learning-based technique for detecting lung cancer and providing explanations of the predictions. This technique is based on a convolutional neural network and XGBoost. XGBoost is used for class label prediction after “DeepXplainer” has automatically learned the features of the input using its many convolutional layers. For providing explanations or explainability of the predictions, an explainable artificial intelligence method known as “SHAP” is implemented. ResultsThe open-source “Survey Lung Cancer” dataset was processed using this method. On multiple parameters, including accuracy, sensitivity, F1-score, etc., the proposed method outperformed the existing methods. The proposed method obtained an accuracy of 97.43%, a sensitivity of 98.71%, and an F1-score of 98.08. After the model has made predictions with this high degree of accuracy, each prediction is explained by implementing an explainable artificial intelligence method at both the local and global levels. ConclusionsA deep learning-based classification model for lung cancer is proposed with three primary components: one for feature learning, another for classification, and a third for providing explanations for the predictions made by the proposed hybrid (ConvXGB) model. The proposed “DeepXplainer” has been evaluated using a variety of metrics, and the results demonstrate that it outperforms the current benchmarks. Providing explanations for the predictions, the proposed approach may help doctors in detecting and treating lung cancer patients more effectively.

Deep learning-based cancer disease classification using Gene Expression Data

&lt;p&gt;Cancer disease caused a major death in worldwide. To prevent this, various cancer classification approaches are employed, which are mostly relied on clinical characteristics and histopathological characteristics. Deep learning-based classification models are very effective and accurate. As a result, this research established the Adam-based Deep Quantum Neural Network, which is the optimal deep learning-based cancer classification method. The information on gene expression is used to classify cancer. Using the Box-Cox transformation, which converts the data into a legible format, the data transformation procedure is carried out. Utilizing information gain, the features are chosen in order to choose the proper gene expression. Additionally, Deep QNN is used to classify cancer, and for better classification, which is trained via Adam optimization. The experimental result shows the developed model provide better classification result with respect to accuracy, true positive rate and true negative rate of 94.91%, 95.59% and 95.4%.&lt;/p&gt;

An Automatic Non-Destructive External and Internal Quality Evaluation of Mango Fruits based on Color and X-ray Imaging with Machine Learning and Deep Learning Based Classification Models

Quality evaluation of food products, agricultural produce to be specific, has gained momentum from past few decades due to the increased awareness among consumers across the world. This has resulted in the increased emphasis on the development and use of quality assessment techniques in food industry. Moreover, there is a need to automate the quality monitoring of agricultural produce like fruits and vegetables which is otherwise done manually in developing countries hence labor intensive, time consuming and subjective in nature. This paper presents an empirical analysis to build a rapid, robust, real-time, non-destructive computer vision based quality assessment model for mango fruits. The work employs the automatic disease classification of mango fruits based on machine and deep learning models. Firstly, the dataset of colored mango fruits images with 2279 images falling into three classes and another dataset of soft X-ray images of mango fruits with 572 images belonging to two quality classes are developed for detecting external and internal defects, respectively. The multilayer perceptron neural network (MLP NN) with two hidden layers, which may be considered as the starting point for deep learning technique, is proposed as machine learning model to classify the color images of mango fruits into one of three external quality classes with 95.1% accuracy and also to classify the soft X-ray images into two internal quality classes with 97.5% accuracy. In order to step out of feature engineering, actual deep learning convolutional neural network (CNN) models, a customized CNN model and pre-trained CNN models, VGGNet (VGG16) and DenseNet121 were also explored for mango disease classification. The maximum validation accuracy of custom CNN was found to be with 91.52% and 98.7% for color and augmented X-ray images, respectively. The classification accuracy of pre-trained models were found to be reasonably good for the color images but exhibited high variability in results and made it difficult to draw a general conclusion for the proposed datasets. However, the proposed MLP NN model based on few basic intensity and geometric features and also the proposed customized CNN model were found to be the best models and they outperform the state of the art reported in the literature.

Multivariate Process Control Chart Pattern Classification Using Multi-Channel Deep Convolutional Neural Networks

Statistical process control (SPC) charts are commonly used to monitor quality characteristics in manufacturing processes. When monitoring two or more related quality characteristics simultaneously, multivariate T2 control charts are often employed. Like univariate control charts, control chart pattern recognition (CCPR) plays a crucial role in multivariate SPC. The presence of non-random patterns in T2 control charts indicates that a process is influenced by one or more assignable causes and that corrective actions should be taken. In this study, we developed a deep learning-based classification model for recognizing control chart patterns in multivariate processes. To address the problem of the insufficient representation of one-dimensional (1D) data, we explore the advantages of using two-dimensional (2D) image data obtained from a threshold-free recurrence plot. A multi-channel deep convolutional neural network (MCDCNN) model was developed to incorporate both 1D and 2D representations of control chart data. This model was tested on multivariate processes with different covariance matrices and compared with other traditional algorithms. Moreover, the effects of imbalanced datasets and dataset size on classification performance were analyzed. Simulation studies revealed that the developed MCDCNN model outperforms other techniques in identifying multivariate non-random patterns. For the most significant one, our proposed MCDCNN method achieved a 10% improvement over traditional methods. The overall results suggest that the developed MCDCNN model can be beneficial for intelligent SPC.

Deep learning-based classification model for GPR151 activator activity prediction

BackgroundGPR151 is a kind of protein belonging to G protein-coupled receptor family that is closely associated with a variety of physiological and pathological processes.The potential use of GPR151 as a therapeutic target for the management of metabolic disorders has been demonstrated in several studies, highlighting the demand to explore its activators further. Activity prediction serves as a vital preliminary step in drug discovery, which is both costly and time-consuming. Thus, the development of reliable activity classification model has become an essential way in the process of drug discovery, aiming to enhance the efficiency of virtual screening.ResultsWe propose a learning-based method based on feature extractor and deep neural network to predict the activity of GPR151 activators. We first introduce a new molecular feature extraction algorithm which utilizes the idea of bag-of-words model in natural language to densify the sparse fingerprint vector. Mol2vec method is also used to extract diverse features. Then, we construct three classical feature selection algorithms and three types of deep learning model to enhance the representational capacity of molecules and predict activity label by five different classifiers. We conduct experiments using our own dataset of GPR151 activators. The results demonstrate high classification accuracy and stability, with the optimal model Mol2vec-CNN significantly improving performance across multiple classifiers. The svm classifier achieves the best accuracy of 0.92 and F1 score of 0.76 which indicates promising applications for our method in the field of activity prediction.ConclusionThe results suggest that the experimental design of this study is appropriate and well-conceived. The deep learning-based feature extraction algorithm established in this study outperforms traditional feature selection algorithm for activity prediction. The model developed can be effectively utilized in the pre-screening stage of drug virtual screening.

Real-Time Object Detection Using SSD MobileNet Model of Machine Learning

This research paper focuses on the application of computer vision techniques using Python and OpenCV for image analysis and interpretation. The main objective is to develop a system capable of performing various tasks such as object detection, recognition, and image processing. The project employs a combination of traditional computer vision algorithms and deep learning models to achieve accurate and efficient results. The research paper begins with essential preprocessing steps, including image acquisition, resizing, and noise reduction. Feature extraction techniques are utilized to capture relevant information from images, followed by object detection using methods like Haar cascades or deep learning-based approaches such as YOLO. Object recognition is achieved through feature matching or deep learning-based classification models. Furthermore, image processing techniques, including image enhancement, segmentation, and filtering, are applied to improve image quality and extract meaningful information. The system is implemented using Python programming language, leveraging the powerful OpenCV library for various computer vision tasks.

TS-GAN: Time-series GAN for Sensor-based Health Data Augmentation

Deep learning has achieved significant success on intelligent medical treatments, such as automatic diagnosis and analysis of medical data. To train an automatic diagnosis system with high accuracy and strong robustness in healthcare, sufficient training data are required when using deep learning-based methods. However, given that the data collected by sensors that are embedded in medical or mobile devices are inadequate, it is challenging to train an effective and efficient classification model with state-of-the-art performance. Inspired by generative adversarial networks (GANs), we propose TS-GAN, a Time-series GAN architecture based on long short-term memory (LSTM) networks for sensor-based health data augmentation, thereby improving the performance of deep learning-based classification models. TS-GAN aims to learn a generative model that creates time-series data with the same space and time dependence as the real data. Specifically, we design an LSTM-based generator for creating realistic data and an LSTM-based discriminator for determining how similar the generated data are to real data. In particular, we design a sequential-squeeze-and-excitation module in the LSTM-based discriminator to better understand space dependence of real data, and apply the gradient penalty originated from Wasserstein GANs in the training process to stabilize the optimization. We conduct comparative experiments to evaluate the performance of TS-GAN with TimeGAN, C-RNN-GAN and Conditional Wasserstein GANs through discriminator loss, maximum mean discrepancy, visualization methods and classification accuracy on health datasets of ECG_200, NonInvasiveFatalECG_Thorax1, and mHealth, respectively. The experimental results show that TS-GAN exceeds other state-of-the-art time-series GANs in almost all the evaluation metrics, and the classifier trained on synthetic datasets generated by TS-GAN achieves the highest classification accuracy of 97.50% on ECG_200, 94.12% on NonInvasiveFatalECG_Thorax1, and 98.12% on mHealth, respectively.

Deep learning-based construction equipment operators’ mental fatigue classification using wearable EEG sensor data

Operator attention failure due to mental fatigue during extended equipment operations is a common cause of equipment-related accidents that result in catastrophic injuries and fatalities. As a result, tracking operators' mental fatigue is critical to reducing equipment-related accidents on construction sites. Previously, several strategies aimed at recognizing mental fatigue with adequate accuracy, such as machine learning utilizing EEG-based wearable sensing systems, have been proposed. However, the ability to track operators’ mental fatigue for its implementation on an actual construction site is still an issue. For instance, the mobility and systemic instability of EEG sensors necessitate their application in laboratory settings rather than on actual construction sites. Furthermore, while the machine learning classifiers achieved acceptable accuracy, their input is limited to manually developed EEG features, which may compromise the models’ performance on real construction sites. Accordingly, the current research proposes the viability of a construction site strategy that uses flexible headband-based sensors for acquiring raw EEG data and deep learning networks to recognize operators' mental fatigue. To serve this purpose, a one-hour excavator operation by fifteen operators was conducted on a construction site. The NASA-TLX score was used as the ground truth of mental fatigue, and brain activity patterns were recorded using a wearable EEG sensor. The raw EEG data was then used to develop deep learning-based classification models. Finally, the performance of deep learning models, i.e., long short-term memory, bidirectional LSTM, and one-dimensional convolutional networks, was investigated using accuracy, precision, recall, specificity, and an F1-score. The findings indicate that the Bi-LSTM model outperforms the other deep learning models with a high accuracy of 99.941% and F1-score between 99.917% and 99.993%. These findings demonstrate the feasibility of applying the Bi-LSTM model and contribute to wearable sensor-based mental fatigue recognition and classification, thus enhancing on-site health and safety operations.