Search

Social media (SM) based surveillance systems, combined with machine learning (ML) and deep learning (DL) techniques, have shown potential for early detection of epidemic outbreaks. This review discusses the current state of SM-based surveillance methods for early epidemic outbreaks and the role of ML and DL in enhancing their performance. Since, every year, a large amount of data related to epidemic outbreaks, particularly Twitter data is generated by SM. This paper outlines the theme of SM analysis for tracking health-related issues and detecting epidemic outbreaks in SM, along with the ML and DL techniques that have been configured for the detection of epidemic outbreaks. DL has emerged as a promising ML technique that adapts multiple layers of representations or features of the data and yields state-of-the-art extrapolation results. In recent years, along with the success of ML and DL in many other application domains, both ML and DL are also popularly used in SM analysis. This paper aims to provide an overview of epidemic outbreaks in SM and then outlines a comprehensive analysis of ML and DL approaches and their existing applications in SM analysis. Finally, this review serves the purpose of offering suggestions, ideas, and proposals, along with highlighting the ongoing challenges in the field of early outbreak detection that still need to be addressed. [ABSTRACT FROM AUTHOR]

Urban air mobility, also known as UAM, is currently being researched in a variety of metropolitan regions throughout the world as a potential new mode of transport for travelling shorter distances inside a territory. In this article, we investigate whether or not the market can back the necessary financial commitments to deploy UAM. A challenge in defining and addressing a critical phase of such guidance is called a demand forecast problem. To achieve this goal, a deep learning model for forecasting temporal data is proposed. This model is used to find and study the scientific issues involved. A benchmark dataset of 150,000 records was used for this purpose. Our experiments used different state-of-the-art DL models: LSTM, GRU, and Transformer for UAM demand prediction. The transformer showed a high performance with an RMSE of 0.64, allowing decision-makers to analyze the feasibility and viability of their investments. [ABSTRACT FROM AUTHOR]

Software vulnerabilities pose a significant threat to system security, necessitating effective automatic detection methods. Current techniques face challenges such as dependency issues, language bias, and coarse detection granularity. This study presents a novel deep learning-based vulnerability detection system for Java code. Leveraging hybrid feature extraction through graph and sequence-based techniques enhances semantic and syntactic understanding. The system utilizes control flow graphs (CFG), abstract syntax trees (AST), program dependencies (PD), and greedy longest-match first vectorization for graph representation. A hybrid neural network (GCN-RFEMLP) and the pre-trained CodeBERT model extract features, feeding them into a quantum convolutional neural network with self-attentive pooling. The system addresses issues like long-term information dependency and coarse detection granularity, employing intermediate code representation and inter-procedural slice code. To mitigate language bias, a benchmark software assurance reference dataset is employed. Evaluations demonstrate the system's superiority, achieving 99.2% accuracy in detecting vulnerabilities, outperforming benchmark methods. The proposed approach comprehensively addresses vulnerabilities, including improper input validation, missing authorizations, buffer overflow, cross-site scripting, and SQL injection attacks listed by common weakness enumeration (CWE). [ABSTRACT FROM AUTHOR]

A brain tumor is an unnatural expansion of brain cells that can't be stopped, making it one of the deadliest diseases of the nervous system. The brain tumor segmentation for its earlier diagnosis is a difficult task in the field of medical image analysis. Earlier, segmenting brain tumors was done manually by radiologists but that requires a lot of time and effort. Inspite of this, in the manual segmentation there was possibility of making mistakes due to human intervention. It has been proved that deep learning models can outperform human experts for the diagnosis of brain tumor in MRI images. These algorithms employ a huge number of MRI scans to learn the difficult patterns of brain tumors to segment them automatically and accurately. Here, an encoder-decoder based architecture with deep convolutional neural network is proposed for semantic segmentation of brain tumor in MRI images. The proposed method focuses on the image downsampling in the encoder part. For this, an intelligent LinkNet-34 model with EfficientNetB7 encoder based semantic segmentation model is proposed. The performance of LinkNet-34 model is compared with other three models namely FPN, U-Net, and PSPNet. Further, the performance of EfficientNetB7 used as encoder in LinkNet-34 model has been compared with three encoders namely ResNet34, MobileNet_V2, and ResNet50. After that, the proposed model is optimized using three different optimizers such as RMSProp, Adamax and Adam. The LinkNet-34 model has outperformed with EfficientNetB7 encoder using Adamax optimizer with the value of jaccard index as 0.89 and dice coefficient as 0.915. [ABSTRACT FROM AUTHOR]

Forecasting rainfall is crucial to the well-being of individuals and is significant everywhere in the world. It contributes to reducing the disastrous effects of floods on agriculture, human life, and socioeconomic systems. This study discusses the challenges of effectively forecasting rainfall and floods and the necessity of combining data with flood channel mathematical modelling to forecast floodwater levels and velocities. This research focuses on leveraging historical meteorological data to find trends using machine learning and deep learning approaches to estimate rainfall. The Bangladesh Meteorological Department provided the data for the study, which also uses eight machine learning algorithms. The performance of the machine learning models is examined using evaluation measures like the R2 score, root mean squared error and validation loss. According to this research's findings, polynomial regression, random forest regression, and long short-term memory (LSTM) had the highest performance levels. Random forest and polynomial regression have an R2 value of 0.76, while LSTM has a loss value of 0.09, respectively. [ABSTRACT FROM AUTHOR]

Bone marrow (BM) is an essential part of the hematopoietic system, which generates all of the body's blood cells and maintains the body's overall health and immune system. The classification of bone marrow cells is pivotal in both clinical and research settings because many hematological diseases, such as leukemia, myelodysplastic syndromes, and anemias, are diagnosed based on specific abnormalities in the number, type, or morphology of bone marrow cells. There is a requirement for developing a robust deep-learning algorithm to diagnose bone marrow cells to keep a close check on them. This study proposes a framework for categorizing bone marrow cells into seven classes. In the proposed framework, five transfer learning models—DenseNet121, EfficientNetB5, ResNet50, Xception, and MobileNetV2—are implemented into the bone marrow dataset to classify them into seven classes. The best-performing DenseNet121 model was fine-tuned by adding one batch-normalization layer, one dropout layer, and two dense layers. The proposed fine-tuned DenseNet121 model was optimized using several optimizers, such as AdaGrad, AdaDelta, Adamax, RMSprop, and SGD, along with different batch sizes of 16, 32, 64, and 128. The fine-tuned DenseNet121 model was integrated with an attention mechanism to improve its performance by allowing the model to focus on the most relevant features or regions of the image, which can be particularly beneficial in medical imaging, where certain regions might have critical diagnostic information. The proposed fine-tuned and integrated DenseNet121 achieved the highest accuracy, with a training success rate of 99.97% and a testing success rate of 97.01%. The key hyperparameters, such as batch size, number of epochs, and different optimizers, were all considered for optimizing these pre-trained models to select the best model. This study will help in medical research to effectively classify the BM cells to prevent diseases like leukemia. [ABSTRACT FROM AUTHOR]

Glomeruli are interconnected capillaries in the renal cortex that are responsible for blood filtration. Damage to these glomeruli often signifies the presence of kidney disorders like glomerulonephritis and glomerulosclerosis, which can ultimately lead to chronic kidney disease and kidney failure. The timely detection of such conditions is essential for effective treatment. This paper proposes a modified UNet model to accurately detect glomeruli in whole-slide images of kidney tissue. The UNet model was modified by changing the number of filters and feature map dimensions from the first to the last layer to enhance the model's capacity for feature extraction. Moreover, the depth of the UNet model was also improved by adding one more convolution block to both the encoder and decoder sections. The dataset used in the study comprised 20 large whole-side images. Due to their large size, the images were cropped into 512 × 512-pixel patches, resulting in a dataset comprising 50,486 images. The proposed model performed well, with 95.7% accuracy, 97.2% precision, 96.4% recall, and 96.7% F1-score. These results demonstrate the proposed model's superior performance compared to the original UNet model, the UNet model with EfficientNetb3, and the current state-of-the-art. Based on these experimental findings, it has been determined that the proposed model accurately identifies glomeruli in extracted kidney patches. [ABSTRACT FROM AUTHOR]

Knee osteoarthritis (OA) is a chronic degenerative joint disease affecting millions worldwide, particularly those over 60. It is a significant cause of disability and can impact an individual’s quality of life. The condition occurs when the cartilage in the knee joint wears away over time, leading to bone-onbone contact, which can result in pain, stiffness, swelling, and decreased range of motion. Deep neural networks, especially convolutional neural networks (CNN), are powerful tools in medical applications such as diagnosis and detection. This research proposes a CNN model to classify knee osteoarthritis into five categories using x-ray images. These classes are labeled: Minimal, Healthy, Moderate, Doubtful, and Severe. Furthermore, the proposed CNN model has been compared with two pre-trained transfer learning models: Xception and InceptionResNet V2. These models were evaluated based on precision, recall, F1 score, and accuracy. The results showed that although all three models performed very well, the proposed model outperformed both transfer learning models with 98% accuracy. It also achieved the highest values for other parameters such as precision, recall, and F1 score. The proposed model has several potential applications in clinical practice, such as assisting doctors in accurately classifying knee osteoarthritis severity levels by analyzing single X-ray images. [ABSTRACT FROM AUTHOR]

Acute Lymphocytic Leukemia is a type of cancer that occurs when abnormal white blood cells are produced in the bone marrow which do not function properly, crowding out healthy cells and weakening the immunity of the body and thus its ability to resist infections. It spreads quickly in children's bodies, and if not treated promptly it may lead to death. The manual detection of this disease is a tedious and slow task. Machine learning and deep learning techniques are faster than manual detection and more accurate. In this paper, a deep feature selection-based approach ResRandSVM is proposed for the detection of Acute Lymphocytic Leukemia in blood smear images. The proposed approach uses seven deep-learning models: ResNet152, VGG16, DenseNet121, MobileNetV2, InceptionV3, EfficientNetB0 and ResNet50 for deep feature extraction from blood smear images. After that, three feature selection methods are used to extract valuable and important features: analysis of variance (ANOVA), principal component analysis (PCA), and Random Forest. Then the selected feature map is fed to four different classifiers, Adaboost, Support Vector Machine, Artificial Neural Network and Naïve Bayes models, to classify the images into leukemia and normal images. The model performs best with a combination of ResNet50 as a feature extractor, Random Forest as feature selection and Support Vector Machine as a classifier with an accuracy of 0.900, precision of 0.902, recall of 0.957 and F1-score of 0.929. [ABSTRACT FROM AUTHOR]

Pneumonia has been directly responsible for a huge number of deaths all across the globe. Pneumonia shares visual features with other respiratory diseases, such as tuberculosis, which can make it difficult to distinguish between them. Moreover, there is significant variability in the way chest X-ray images are acquired and processed, which can impact the quality and consistency of the images. This can make it challenging to develop robust algorithms that can accurately identify pneumonia in all types of images. Hence, there is a need to develop robust, data-driven algorithms that are trained on large, high-quality datasets and validated using a range of imaging techniques and expert radiological analysis. In this research, a deep-learning-based model is demonstrated for differentiating between normal and severe cases of pneumonia. This complete proposed system has a total of eight pre-trained models, namely, ResNet50, ResNet152V2, DenseNet121, DenseNet201, Xception, VGG16, EfficientNet, and MobileNet. These eight pre-trained models were simulated on two datasets having 5856 images and 112,120 images of chest X-rays. The best accuracy is obtained on the MobileNet model with values of 94.23% and 93.75% on two different datasets. Key hyperparameters including batch sizes, number of epochs, and different optimizers have all been considered during comparative interpretation of these models to determine the most appropriate model. [ABSTRACT FROM AUTHOR]

The segmentation of lungs from medical images is a critical step in the diagnosis and treatment of lung diseases. Deep learning techniques have shown great promise in automating this task, eliminating the need for manual annotation by radiologists. In this research, a convolution neural network architecture is proposed for lung segmentation using chest X-ray images. In the proposed model, concatenate block is embedded to learn a series of filters or features used to extract meaningful information from the image. Moreover, a transpose layer is employed in the concatenate block to improve the spatial resolution of feature maps generated by a prior convolutional layer. The proposed model is trained using k-fold validation as it is a powerful and flexible tool for evaluating the performance of deep learning models. The proposed model is evaluated on five different subsets of the data by taking the value of k as 5 to obtain the optimized model to obtain more accurate results. The performance of the proposed model is analyzed for different hyper-parameters such as the batch size as 32, optimizer as Adam and 40 epochs. The dataset used for the segmentation of disease is taken from the Kaggle repository. The various performance parameters such as accuracy, IoU, and dice coefficient are calculated, and the values obtained are 0.97, 0.93, and 0.96, respectively. [ABSTRACT FROM AUTHOR]

Brain tumor diagnosis at an early stage can improve the chances of successful treatment and better patient outcomes. In the biomedical industry, non-invasive diagnostic procedures, such as magnetic resonance imaging (MRI), can be used to diagnose brain tumors. Deep learning, a type of artificial intelligence, can analyze MRI images in a matter of seconds, reducing the time it takes for diagnosis and potentially improving patient outcomes. Furthermore, an ensemble model can help increase the accuracy of classification by combining the strengths of multiple models and compensating for their individual weaknesses. Therefore, in this research, a weighted average ensemble deep learning model is proposed for the classification of brain tumors. For the weighted ensemble classification model, three different feature spaces are taken from the transfer learning VGG19 model, Convolution Neural Network (CNN) model without augmentation, and CNN model with augmentation. These three feature spaces are ensembled with the best combination of weights, i.e., weight1, weight2, and weight3 by using grid search. The dataset used for simulation is taken from The Cancer Genome Atlas (TCGA), having a lower-grade glioma collection with 3929 MRI images of 110 patients. The ensemble model helps reduce overfitting by combining multiple models that have learned different aspects of the data. The proposed ensemble model outperforms the three individual models for detecting brain tumors in terms of accuracy, precision, and F1-score. Therefore, the proposed model can act as a second opinion tool for radiologists to diagnose the tumor from MRI images of the brain. [ABSTRACT FROM AUTHOR]

The human gastrointestinal (GI) tract is an important part of the body. According to World Health Organization (WHO) research, GI tract infections kill 1.8 million people each year. In the year 2019, almost 5 million individuals were detected with gastrointestinal disease. Radiation therapy has the potential to improve cure rates in GI cancer patients. Radiation oncologists direct X-ray beams at the tumour while avoiding the stomach and intestines. The current objective is to direct the X-ray beam toward the malignancy while avoiding the stomach and intestines in order to improve dose delivery to the tumour. This study offered a technique for segmenting GI tract organs (small bowel, big intestine, and stomach) to assist radio oncologists to treat cancer patients more quickly and accurately. The suggested model is a U-Net model designed from scratch and used for the segmentation of a small size of images to extract the local features more efficiently. Furthermore, in the proposed model, six transfer learning models were employed as the backbone of the U-Net topology. The six transfer learning models used are Inception V3, SeResNet50, VGG19, DenseNet121, InceptionResNetV2, and EfficientNet B0. The suggested model was analysed with model loss, dice coefficient, and IoU. The results specify that the suggested model outperforms all transfer learning models, with performance parameter values as 0.122 model loss, 0.8854 dice coefficient, and 0.8819 IoU. [ABSTRACT FROM AUTHOR]

Clean air is necessary for leading a healthy life. Many respiratory illnesses have their root in the poor quality of air across regions. Due to the tremendous impact of air quality on people's lives, it is essential to devise a mechanism through which air pollutants (PM2.5, NOx, COx, SOx) can be forecasted. However, forecasting air quality and its pollutants is complicated as air quality depends on several factors such as weather, vehicular, and power plant emissions. This aim of this research was to find the impact of weather on PM2.5 concentrations and to forecast the daily and hourly PM2.5 concentration for the next 30 days and 72 h in Pakistan. This forecasting was done through state-of-the-art deep learning and machine learning models such as FbProphet, LSTM, and LSTM encoder–decoder. This research also successfully forecasted the proposed daily and hourly PM2.5 concentration. The LSTM encoder–decoder had the best performance and successfully forecasted PM2.5 concentration with a mean absolute percentage error (MAPE) of 28.2%, 15.07%, and 42.1% daily, and 11.75%, 9.5%, and 7.4% hourly for different cities in Pakistan. This research proves that a data-driven approach is essential for resolving air pollution in Pakistan. [ABSTRACT FROM AUTHOR]

Sentiment analysis taskhas widely been studied for various languages such as English and French. However, Roman Urdu sentiment analysis yet requires more attention from peer-researchers due to the lack of Off-the-Shelf Natural Language Processing (NLP) solutions. The primary objective of this study is to investigate the diverse machine learning methods for the sentiment analysis of Roman Urdu data which is very informal in nature and needs to be lexically normalized. To mitigate this challenge, we propose a fine-tuned Support Vector Machine (SVM) powered by Roman Urdu Stemmer. In our proposed scheme, the corpus data is initially cleaned to remove the anomalies from the text. After initial pre-processing, each user review is being stemmed. The input text is transformed into a feature vector using the bag-of-word model. Subsequently, the SVM is used to classify and detect user sentiment. Our proposed scheme is based on a dictionary based Roman Urdu stemmer. The creation of the Roman Urdu stemmer is aimed at standardizing the text so as to minimize the level of complexity. The efficacy of our proposed model is also empirically evaluated with diverse experimental configurations, so as to fine-tune the hyper-parameters and achieve superior performance. Moreover, a series of experiments are conducted on diverse machine learning and deep learning models to compare the performance with our proposed model. We also introduced the largest dataset on Roman Urdu, i.e., Roman Urdu e-commerce dataset (RUECD), which contains 26K+ user reviews annotated by the group of experts. The RUECD is challenging and the largest dataset available of Roman Urdu. The experiments show that the newly generated dataset is quite challenging and requires more attention from the peer researchers for Roman Urdu sentiment analysis. [ABSTRACT FROM AUTHOR]

High-dimensional medical data makes prediction a complex and difficult task. This study aims at modeling predictive models for medical data. Two datasets of medical data are applied in the study — one online available dataset (Heart Disease data) and another real clinical dataset (Eye Infection Data). A wide range of machine learning algorithms are applied in the modeling stage: Decision Tree, Multilayer Perceptron, Naive Bayesian, Random Forest, and Support Vector Machine. Furthermore, bagging and voting ensemble methods have also been applied with base learners. Both split and cross-validation methods are adopted for the model validation, and well-established evaluation metrics such as accuracy, precision, recall, and F-measure have been considered as evaluation metrics for the predictive models. The method applied for the modeling is comprised of two stages. The first stage uses available features for the predictions. In the second stage, selected features based on positive correlation are used. The adopted method is also for deep learning, especially Convolutional Neural Network (CNN) is applied to analyze the outcomes compared to conventional machine learning algorithms. The experimental results reveal that better predictions are achieved in the second stage. Besides, experiments also indicate split percentage produces better predictive models, and marginally better outcomes are observed in the presence of ensemble methods in comparison with base models. NB outperformed other algorithms with the highest accuracy rate as 88.90%, and MLP obtained 97.50% accuracy for Heart Disease and Eye Infection data, respectively, using 80–20 splits in the second stage. However, the CNN model performed poorly due to the size of the considered datasets. • Two medical datasets are applied in the study: one online available and the other real. • Machine learning algorithms including DT, MLP, NB and RF are applied in the modeling. • The method applied for the modeling is comprised of two stages. • The experimental results reveal that better predictions are achieved in the second stage. • NB outperformed for an online dataset with 88.90% accuracy and MLP with 97.50% for real dataset. [ABSTRACT FROM AUTHOR]

The classification of whole slide images (WSIs) provides physicians with an accurate analysis of diseases and also helps them to treat patients effectively. The classification can be linked to further detailed analysis and diagnosis. Deep learning (DL) has made significant advances in the medical industry, including the use of magnetic resonance imaging (MRI) scans, computerized tomography (CT) scans, and electrocardiograms (ECGs) to detect life-threatening diseases, including heart disease, cancer, and brain tumors. However, more advancement in the field of pathology is needed, but the main hurdle causing the slow progress is the shortage of large-labeled datasets of histopathology images to train the models. The Kimia Path24 dataset was particularly created for the classification and retrieval of histopathology images. It contains 23,916 histopathology patches with 24 tissue texture classes. A transfer learning-based framework is proposed and evaluated on two famous DL models, Inception-V3 and VGG-16. To improve the productivity of Inception-V3 and VGG-16, we used their pre-trained weights and concatenated these with an image vector, which is used as input for the training of the same architecture. Experiments show that the proposed innovation improves the accuracy of both famous models. The patch-to-scan accuracy of VGG-16 is improved from 0.65 to 0.77, and for the Inception-V3, it is improved from 0.74 to 0.79. [ABSTRACT FROM AUTHOR]