DL Approach Research Articles

AUTOMATED GESTURE RECOGNITION USING APPLIED LINGUISTICS WITH DATA-DRIVEN DEEP LEARNING FOR ARABIC SPEECH TRANSLATION

Gesture recognition for Arabic speech translation includes developing advanced technologies that correctly translate body and hand movements corresponding to Arabic sign language (ArSL) into spoken Arabic. This leverages machine learning and computer vision techniques in complex systems simulation platforms to scrutinize the gestures utilized in ArSL, detecting mild differences in facial expressions, hand shapes, and movements. Sign Language Recognition (SLR) is paramount in assisting communication for the Deaf and Hard-of-Hearing communities. It includes using vision-based methods and Surface Electromyography (sEMG) signals. The sEMG signal is crucial for recognizing hand gestures and capturing muscular activities in sign language. Researchers have comprehensively shown the capability of EMG signals to approach specific details, mainly in classifying hand gestures. This progression is a stimulating feature in extracting the interpretation and recognition of sign languages and investigating the phonology of signed language. Leveraging machine learning algorithms and signal processing techniques in complex systems simulation platforms, researchers aim to extract relevant traits from the sEMG signals that correspond to different ArSL gestures. This study introduces an Enhanced Dwarf Mongoose Algorithm with a Deep Learning-Driven Arabic Sign Language Detection (EDMODL-ASLD) technique on sEMG data. In the initial phase, the presented EDMODL-ASLD model is subjected to data preprocessing to change the input sEMG data into an attuned format. In the next stage, feature extraction with fractal theories is used to gather relevant and nonredundant data from the EMG window to construct a feature vector. In this study, the absolute envelope (AE), energy (E), root-mean square (RMS), standard deviation (STD), and mean absolute value (MAV) are the five time-domain extracted features for the EMG window observation. Meanwhile, the dilated convolutional long short-term memory (ConvLSTM) technique is used to identify distinct sign languages. To improve the results of the dilated ConvLSTM model, the hyperparameter selection process is executed using the EDMO model. To illustrate the significance of the EDMODL-ASLD technique, a brief experimental validation is made on the Arabic SLR database. The experimental validation of the EDMODL-ASLD technique portrayed a superior accuracy value of 96.47% over recent DL approaches.

Impact of Machine Learning Integration in Qur’anic Studies

The advancement in the field of computer science, especially in machine learning (ML), represents a flourishing innovation that carries great importance in the domain of education. The beneficial impact of ML can also be observed in the realm of Qur’anic studies, particularly in Arabic text recognition and recitation analysis. This paper presents a comprehensive analysis of 34+ published scholarly articles devoted to Qur’anic studies. This work explores the convergence of machine learning methodologies and Qur’anic studies, examining the innovative applications and methodologies for Arabic text and voice classification. The fusion of ML algorithms makes the work easy and accurate to analyze, interpret, and extract valuable insights from the sacred text. Subsequently, we delve deeper into the emergent field of ML algorithms like k-NN, ANN, BLSTM, MFCC, SVM, NB and DL approaches have been adapted for Qur’anic texts classification, recitation and recitation analysis on accuracy, speed, class recognition, response rate and biasness benchmark. This work covers a diverse range of applications, including automated Qur’anic exegesis and analysis of usage of Ahkam Al-Tajweed. The main contribution of the work is to provide insight into how ML facilitates in Arabic and Kufic textual analysis, linguistic subtleties, and thematic structures of the Qur’anic text. Using the deep learning approaches, the reciters, recitation style and of the Quranic text has also explained in the work.

An Optimized Tuning Based Hybrid DL Approach For Lung Cancer Classification

An Optimized Tuning Based Hybrid DL Approach For Lung Cancer Classification

Deep Hybrid Model for Pest Detection: IoT‐UAV‐Based Smart Agriculture System

ABSTRACTModern technology is revolutionising traditional farming processes by introducing new and streamlined approaches. Despite these advancements, challenges such as disease identification, insect detection and weather forecasting persist. To address these issues, this work proposes a DHMPD‐based IoT‐UAV smart agriculture system focused on pest detection. The method involves several stages: data acquisition, preprocessing, data augmentation, segmentation, feature extraction and classification. During data acquisition, a ‘Pest data set’ is collected. Preprocessing includes Z‐score normalisation to produce better‐normalised images. Data augmentation involves rotating images to create different orientations. The segmentation stage uses an updated HDBSCAN process, which improves the distance calculation between pixels using hybridised Euclidean and Minkowski distances. Feature extraction retrieves various features from segmented images, including modified MBP features, colour‐based features and shape‐based features. After feature extraction, the classification phase is performed by a hybrid technique with DL approaches such as improved DBN and LSTM approaches. Finally, classification results are averaged to predict pest detection accurately. The approach's effectiveness is evaluated through various assessments, aiming to overcome current limitations and enhance smart agriculture systems. The proposed DHMPD method was compared with state‐of‐the‐art approaches and traditional classifiers, achieving a maximum accuracy of 0.936, outperforming conventional methods in accurately detecting pests. Hence, the proposed work holds immense promise to advance the capabilities of smart agriculture systems, offering practical solutions that can benefit farmers, agricultural researchers and industries involved in crop management and food production.

2794: New DL approach for dysphagia analysis in videofluoroscopy swallowing studies in HN cancer patients

2794: New DL approach for dysphagia analysis in videofluoroscopy swallowing studies in HN cancer patients

Enhancing personalized gene expression prediction from DNA sequences using genomic foundation models

Artificial intelligence/deep learning (AI/DL) models that predict molecular phenotypes like gene expression directly from DNA sequences have recently emerged. While these models have proven effective at capturing the variation across genes, their ability to explain inter-individual differences has been limited. We hypothesize that the performance gap can be narrowed through the use of pre-trained embeddings from the Nucleotide Transformer, a large foundation model trained on 3,000+ genomes. We train a transformer model using the pre-trained embeddings and compare its predictive performance to Enformer, the current state-of-the-art model, using genotype and expression data from 290 individuals. Our model significantly outperforms Enformer in terms of correlation across individuals and narrows the performance gap with an elastic net regression approach using just the genetic variants as predictors. Although simple regression models have their advantages in personalized prediction tasks, DL approaches based on foundation models pre-trained on diverse genomes have unique strengths in flexibility and interpretability. With further methodological and computational improvements with more training data, these models may someday predict molecular phenotypes from DNA sequences with accuracy surpassing that of regression-based approaches. Our work demonstrates the potential for large pre-trained AI/DL models to advance functional genomics.

EEG workload estimation and classification: a systematic review.

Electroencephalography (EEG) has evolved into an indispensable instrument for estimating cognitive workload in various domains. ML and DL techniques have been increasingly employed to develop accurate workload estimation and classification models based on EEG data. The goal of this systematic review is to compile the body of research on EEG workload estimation and classification using ML and DL approaches. The PRISMA procedures were followed in conducting the review, searches were conducted through databases at SpringerLink, ACM Digital Library, IEEE Explore, PUBMED, and Science Direct from the beginning to the end of February 16, 2024. Studies were selected based on predefined inclusion criteria. Data were extracted to capture study design, participant demographics, EEG features, ML/DL algorithms, and reported performance metrics. Out of the 125 items that emerged, 33 scientific papers were fully evaluated. The study designs, participant demographics, and EEG workload measurement and categorization techniques used in the investigations differed. SVM, CNN, and hybrid networks are examples of ML and DL approaches that were often used. Analyzing the accuracy scores achieved by different ML/DL models. Furthermore, a relationship was noted between sample frequency and model accuracy, with higher sample frequencies generally leading to improved performance. The percentage distribution of ML/DL methods revealed that SVMs, CNNs, and RNNs were the most commonly utilized techniques, reflecting their robustness in handling EEG data. The comprehensive review emphasizes how ML may be used to identify mental workload across a variety of disciplines using EEG data. Optimizing practical applications requires multimodal data integration, standardization efforts, and real-world validation studies. These systems will also be further improved by addressing ethical issues and investigating new EEG properties, which will improve human-computer interaction and performance assessment.

Analysis of 12-lead ECGs for SARS-CoV-2 detection using deep learning techniques

AbstractThe spread of the COVID-19 pandemic is expected to be uncontrollable by 2020. The main precautions to avoid virus spread have been the introduction of surgical masks or FFP2, sanitization of the hands, and maintaining social distancing. Due to their reliability, molecular tampons are the main detection and prevention methods known as the “Gold Standard”. However, these methods can be particularly uncomfortable. In this case, the analysis of electrocardiogram traces appears to be an alternative method for detecting COVID-19. The dataset used is made up of 1937 images from a study conducted in Pakistan that were preprocessed to train six different neural networks, including MobileNetV2, ResNet-18, ResNet-50, AlexNet, SqueezeNet, and an ad hoc defined neural network. The results show high classification performance, with an accuracy close to 98.94%, as reached by the Resnet-18 network. Moreover, significant attention was devoted to analyzing confusion matrices, revealing the capacity of the networks to identify distinctive features indicative of COVID-19 within ECG data. Finally, it is suggested that in nearly all experiments, including those with low performance, COVID-19 patients are correctly classified, further enhancing the diagnostic potential of ECGs data and DL approach.

ML & DL approach for Handwritten Character Recognition for Gujarati Characters

ML & DL approach for Handwritten Character Recognition for Gujarati Characters

Comprehensive Review on Plant Disease Detection and Identification

Image Agriculture is a major contributor to financial development and is the main source of income in many countries. The primary goal of disease prevention and treatment, notwithstanding the different challenges farmers face, including plant diseases, is to precisely identify and evaluate the disease while the plant is still growing. Object recognition and classification in images has been made possible by rapid advancements in deep learning, or DL, techniques. Recently, DL approaches have been applied to farming and other agricultural applications and have showed potential in a variety of areas. These plants are commonly affected by diseases such as Septoria leaf spots, bacterial spots, late plant blight, and curled yellow leaves. Our work presents a hybrid approach to early illness detection that applies region-based full convolutional networks (RFCN) for early disease identification and region-based convolutional neural networks (RCNN) for diseases classification. We construct a network model based on the EfficientNetB7 model that optimizes at several scales via dilated convolution. The network is constructed layer by layer, top-down, with several optimizations. Real-time crop images taken from crop fields are used in the experiment, and Efficient Net B7 and hybrid convolution models are used. When the suggested ensemble model is implemented in Python, the outcome demonstrates that it has high accuracy and low loss when compared to other current methodologies.

A Review Work: Human Action Recognition in Video Surveillance Using Deep Learning Techniques

Despite being extensively used in numerous uses, precise and effective human activity identification continues to be an interesting research issue in the area of vision for computers. Currently, a lot of investigation is being done on themes like pedestrian activity recognition and ways to recognize people's movements employing depth data, 3D skeletal data, still picture data, or strategies that utilize spatiotemporal interest points. This study aims to investigate and evaluate DL approaches for detecting human activity in video. The focus has been on multiple structures for detecting human activities that use DL as their primary strategy. Based on the application, including identifying faces, emotion identification, action identification, and anomaly identification, the human occurrence forecasts are divided into four different subcategories. The literature has been carried several research based on these recognitions for predicting human behavior and activity for video surveillance applications. The state of the art of four different applications' DL techniques is contrasted. This paper also presents the application areas, scientific issues, and potential goals in the field of DL-based human behavior and activity recognition/detection.

Integrating singular value decomposition with deep learning for enhanced travel time estimation in multimodal freight transportation networks

AbstractMultimodal freight transport allows switching among various modes of transportation to efficiently utilize transport facilities. A multimodal transport system incorporates geographical scales from global to local. Travel time estimation in a multi‐modal cargo transportation network is essential for enhancing supply chain (SC) and logistics operations. Accurate travel time prediction is of great importance for cargo transportation, as it enables SC participants to increase logistics efficiency and quality. It requires adequate input data, which can be generated. In recent times, the machine learning (ML) algorithm has been well‐suited to resolve complex and nonlinear relationships in the collected tracking data. This study designs a deep learning‐powered travel time estimation in multimodal freight transportation networks (DLTTE‐MFTN) technique. The goal of the DLTTE‐MFTN technique is to estimate the travel time using a hyperparameter‐tuned ensemble learning approach. To achieve this, the DLTTE‐MFTN method initially undergoes data pre‐processing to convert the input raw data into a useful format. In addition, the singular value decomposition (SVD) model can be applied for feature dimensionality reduction in multimodal transport data, considerably improving travel time prediction. Besides, the DLTTE‐MFTN method estimates travel time using an ensemble of three DL approaches including one‐dimensional convolutional neural network (1D‐CNN), stacked autoencoder (SAE) attention, and recurrent neural network (RNN). Finally, the hyperparameter tuning of the DL models takes place using the whale optimization algorithm (WOA). The performance analysis of the DLTTE‐MFTN method takes place using the Kaggle dataset. The experimental results stated that the DLTTE‐MFTN technique attains superior performance over other ML and DL models.

Dynamic pricing on maximum concurrency for heterogeneous instances using hyperparameter optimization in dueling deep reinforcement learning in a multi-cloud scenario

Users possess the option to rent instances of various sorts, in a variety of regions, and a variety of availability zones, thanks to cloud service carriers like AWS, GCP, and Azure. In the cloud business right now, fixed price models are king when it comes to pricing. However, as the diversity of cloud providers and users grows, this approach is unable to accurately reflect the market’s current needs for cost savings. As a consequence, a dynamic pricing strategy has become a desirable tactic to better handle the erratic cloud demand. In this study, a deep learning model was used to propose a dynamic pricing structure that ensures service providers are treated fairly in a multi-cloud context. The computational optimization of DL approaches can be severely hampered by the requirement for human hyperparameter selection. Traditional automated solutions to this issue have inadequate durability or fail in specific circumstances. To choose the hyper-parameters in the Dueling Deep Q-Network (DDQN), the hybrid DL approach in this study uses the concept-based wild horse optimization (WHO) method. A community of untamed horses is evolved, and the fitness of the population is evaluated concurrently to estimate the optimum hyper-parameters. The plan changes the price appropriately to promote the use of underutilized resources and discourage the use of overutilized resources. The evaluation’s findings demonstrated that the suggested strategy can lower end-user costs while conducting compute- and data-intensive activities in a multi-cloud environment. The research was concluded by comparing current models after the results were analyzed using various performance indicators.

Chest X-ray image classification using transfer learning and hyperparameter customization for lung disease diagnosis

ABSTRACT Lung diseases often result in severe damage to the respiratory tract, and lead to a high risk of mortality within a short period of time. DL models based on ViT are considered to have promising advantages over CNN architectures in terms of computational efficiency, and accuracy when trained on large ImageNet datasets. In this study, we present a new DL approach based on the combination of CNN with ViT to improve the efficiency of pneumonia diagnosis using medical images. In the first stage, raw images are passed through a local filter to capture local relations on the inputs. The local filter block includes two convolutional layers with kernel 3 × 3. This local filtering method aims to enhance rich features before being fed into the patching layer of the ViT block. The proposed method is experimented on the benchmark chest X-ray dataset. The proposed method is evaluated and compared to some well-known models, which include ViT, VGG19, Resnet50, Densnet201. Experimental results demonstrated that the proposed approach based on CNN and ViT reaches higher efficiency with about 1% accuracy to the standard ViT model, and about 2% higher with VGG19, Resnet50, Densnet201 and smaller in model architecture.

Improving Equity in Deep Learning Medical Applications with the Gerchberg-Saxton Algorithm

AbstractDeep learning (DL) has gained prominence in healthcare for its ability to facilitate early diagnosis, treatment identification with associated prognosis, and varying patient outcome predictions. However, because of highly variable medical practices and unsystematic data collection approaches, DL can unfortunately exacerbate biases and distort estimates. For example, the presence of sampling bias poses a significant challenge to the efficacy and generalizability of any statistical model. Even with DL approaches, selection bias can lead to inconsistent, suboptimal, or inaccurate model results, especially for underrepresented populations. Therefore, without addressing bias, wider implementation of DL approaches can potentially cause unintended harm. In this paper, we studied a novel method for bias reduction that leverages the frequency domain transformation via the Gerchberg-Saxton and corresponding impact on the outcome from a racio-ethnic bias perspective.

An ensemble approach for classification of tympanic membrane conditions using soft voting classifier

Otitis media is a medical concept that represents a range of inflammatory middle ear disorders. The high costs of medical devices utilized by field experts to diagnose the disease relevant to otitis media prevent the widespread use of these devices. This makes it difficult for field experts to make an accurate diagnosis and increases subjectivity in diagnosing the disease. To solve these problems, there is a need to develop computer-aided middle ear disease diagnosis systems. In this study, a deep learning-based approach is proposed for the detection of OM disease to meet this emerging need. This approach is the first that addresses the performance of a voting ensemble framework that uses Inception V3, DenseNet 121, VGG16, MobileNet, and EfficientNet B0 pre-trained DL models. All pre-trained CNN models used in the proposed approach were trained using the Public Ear Imagery dataset, which has a total of 880 otoscopy images, including different eardrum cases such as normal, earwax plug, myringosclerosis, and chronic otitis media. The prediction results of these models were evaluated with voting approaches to increase the overall prediction accuracy. In this context, the performances of both soft and hard voting ensembles were examined. Soft voting ensemble framework achieved highest performance in experiments with 98.8% accuracy, 97.5% sensitivity, and 99.1% specificity. Our proposed model achieved the highest classification performance so far in the current dataset. The results reveal that our voting ensemble-based DL approach showed quite high performance for the diagnosis of middle ear disease. In clinical applications, this approach can provide a preliminary diagnosis of the patient's condition just before field experts make a diagnosis on otoscopic images. Thus, our proposed approach can help field experts to diagnose the disease quickly and accurately. In this way, clinicians can make the final diagnosis by integrating automatic diagnostic prediction with their experience.

Developments in deep learning for change detection in remote sensing: A review

AbstractDeep learning (DL) algorithms have become increasingly popular in recent years for remote sensing applications, particularly in the field of change detection. DL has proven to be successful in automatically identifying changes in satellite images with varying resolutions. The integration of DL with remote sensing has not only facilitated the identification of global and regional changes but has also been a valuable resource for the scientific community. Researchers have developed numerous approaches for change detection, and the proposed work provides a summary of the most recent ones. Additionally, it introduces the common DL techniques used for detecting changes in satellite photos. The meta‐analysis conducted in this article serves two purposes. Firstly, it tracks the evolution of change detection in DL investigations, highlighting the advancements made in this field. Secondly, it utilizes powerful DL‐based change detection algorithms to determine the best strategy for monitoring changes at different resolutions. Furthermore, the proposed work thoroughly analyzes the performance of several DL approaches used for change detection. It discusses the strengths and limitations of these approaches, providing insights into their effectiveness and areas for improvement. The article also discusses future directions for DL‐based change detection, emphasizing the need for further research and development in this area.

Energy Trading and Optimum Scheduling for Microgrids Using Multiple Agents Based DL Approach

In microgrids, policy decisions about electricity trading and scheduling are of paramount importance. These decisions are influenced by various factors, including the microgrid’s infrastructure, energy production and demand, and the competitiveness of the energy market. However, due to the intricate nature of these elements, determining the optimal strategy for a microgrid can be a challenging task. To address this challenge without the need for complex system modeling, we introduce a novel approach using Multi-Agent Reinforcement Learning (MARL) augmented with an attention mechanism. In this approach, each microgrid is treated as a distinct agent, and through interactions with other agents, they learn how to effectively coordinate the utilization of energy resources and engage in energy trading. To facilitate effective agent training, we employ an attention mechanism that enables agents to selectively focus on pertinent background information. Once trained, each agent can make control decisions based solely on its local knowledge. This not only safeguards the privacy of individual microgrids but also reduces communication overhead, making decentralized control feasible. We implement this approach in MATLAB R2020a to create a simulation environment and assess its performance. Our experimental results indicate that our proposed strategy significantly reduces the operational costs of microgrids in comparison to conventional methods.

Toward Reducing IDS Misclassification Using Hybrid DL and ML Approach

Operation centers often face challenges due to the high rate of misclassifications caused by the lower precision in Intrusion Detection System (IDS) models. Despite several research contributions ranging from machine learning and deep learning techniques aiming to reduce false positives and negatives, researchers and security experts consistently encounter a tradeoff between these two types of errors. This indicates a significant opportunity for further contributions in this field. We propose a hybrid model that combines Recurrent Neural Networks (RNN) feature extraction capabilities with Support Vector Machines (SVM) classification abilities. Our model achieves an impressive accuracy rate of 98.2% and significantly reduces misclassification errors compared to contemporary state-of-the-art models. This work shows the potential of hybrid approaches in improving accuracy and reducing false positive and negative errors.

Exploring Huntington's Disease Diagnosis via Artificial Intelligence Models: A Comprehensive Review.

Huntington's Disease (HD) is a devastating neurodegenerative disorder characterized by progressive motor dysfunction, cognitive impairment, and psychiatric symptoms. The early and accurate diagnosis of HD is crucial for effective intervention and patient care. This comprehensive review provides a comprehensive overview of the utilization of Artificial Intelligence (AI) powered algorithms in the diagnosis of HD. This review systematically analyses the existing literature to identify key trends, methodologies, and challenges in this emerging field. It also highlights the potential of ML and DL approaches in automating HD diagnosis through the analysis of clinical, genetic, and neuroimaging data. This review also discusses the limitations and ethical considerations associated with these models and suggests future research directions aimed at improving the early detection and management of Huntington's disease. It also serves as a valuable resource for researchers, clinicians, and healthcare professionals interested in the intersection of machine learning and neurodegenerative disease diagnosis.