Search Results (11)

Chest X-ray interpretation is essential for diagnosing cardiac and respiratory diseases. This study introduces a deep learning ensemble approach that integrates Convolutional Neural Networks (CNNs), including ResNet-152, VGG19, EfficientNet, and a Vision Transformer (ViT), to enhance diagnostic accuracy. Using the NIH Chest X-ray dataset, the methodology involved comprehensive preprocessing, data augmentation, and model optimization techniques to address challenges such as label imbalance and feature variability. Among the individual models, VGG19 exhibited strong performance with a Hamming Loss of 0.1335 and high accuracy in detecting Edema, while ViT excelled in classifying certain conditions like Hernia. Despite the strengths of individual models, the ensemble meta-model achieved the best overall performance, with a Hamming Loss of 0.1408 and consistently higher ROC-AUC values across multiple diseases, demonstrating its superior capability to handle complex classification tasks. This robust ensemble learning framework underscores its potential for reliable and precise disease detection, offering significant improvements over traditional methods. The findings highlight the value of integrating diverse model architectures to address the complexities of multi-label chest X-ray classification, providing a pathway for more accurate, scalable, and accessible diagnostic tools in clinical practice. Full article

Detecting faulty pipelines in water management systems is crucial for ensuring a reliable supply of clean water. Traditional inspection methods are often time-consuming, costly, and prone to errors. This study introduces an AI-based model utilizing images to detect pipeline defects, focusing on leaks, cracks, and corrosion. The YOLOv8 model is employed for object detection due to its exceptional performance in detecting objects, segmentation, pose estimation, tracking, and classification. By training on a large dataset of labeled images, the model effectively learns to identify visual patterns associated with pipeline faults. Experiments conducted on a real-world dataset demonstrate that the AI-based model significantly outperforms traditional methods in detection accuracy. The model also exhibits robustness to various environmental conditions such as lighting changes, camera angles, and occlusions, ensuring reliable performance in diverse scenarios. The efficient processing time of the model enables real-time fault detection in large-scale water distribution networks implementing this AI-based model offers numerous advantages for water management systems. It reduces dependence on manual inspections, thereby saving costs and enhancing operational efficiency. Additionally, the model facilitates proactive maintenance through the early detection of faults, preventing water loss, contamination, and infrastructure damage. The results from the three conducted experiments indicate that the model from Experiment 1 achieves a commendable mAP50 of 90% in detecting faulty pipes, with an overall mAP50 of 74.7%. In contrast, the model from Experiment 3 exhibits superior overall performance, achieving a mAP50 of 76.1%. This research presents a promising approach to improving the reliability and sustainability of water management systems through AI-based fault detection using image analysis. Full article

The growing problem of unsolicited text messages (smishing) and data irregularities necessitates stronger spam detection solutions. This paper explores the development of a sophisticated model designed to identify smishing messages by understanding the complex relationships among words, images, and context-specific factors, areas that remain underexplored in existing research. To address this, we merge a UCI spam dataset of regular text messages with real-world spam data, leveraging OCR technology for comprehensive analysis. The study employs a combination of traditional machine learning models, including K-means, Non-Negative Matrix Factorization, and Gaussian Mixture Models, along with feature extraction techniques such as TF-IDF and PCA. Additionally, deep learning models like RNN-Flatten, LSTM, and Bi-LSTM are utilized. The selection of these models is driven by their complementary strengths in capturing both the linear and non-linear relationships inherent in smishing messages. Machine learning models are chosen for their efficiency in handling structured text data, while deep learning models are selected for their superior ability to capture sequential dependencies and contextual nuances. The performance of these models is rigorously evaluated using metrics like accuracy, precision, recall, and F1 score, enabling a comparative analysis between the machine learning and deep learning approaches. Notably, the K-means feature extraction with vectorizer achieved 91.01% accuracy, and the KNN-Flatten model reached 94.13% accuracy, emerging as the top performer. The rationale behind highlighting these models is their potential to significantly improve smishing detection rates. For instance, the high accuracy of the KNN-Flatten model suggests its applicability in real-time spam detection systems, but its computational complexity might limit scalability in large-scale deployments. Similarly, while K-means with vectorizer excels in accuracy, it may struggle with the dynamic and evolving nature of smishing attacks, necessitating continual retraining. Full article

Metastatic breast cancer (MBC) continues to be a leading cause of cancer-related deaths among women. This work introduces an innovative non-invasive breast cancer classification model designed to improve the identification of cancer metastases. While this study marks the initial exploration into predicting MBC, additional investigations are essential to validate the occurrence of MBC. Our approach combines the strengths of large language models (LLMs), specifically the bidirectional encoder representations from transformers (BERT) model, with the powerful capabilities of graph neural networks (GNNs) to predict MBC patients based on their histopathology reports. This paper introduces a BERT-GNN approach for metastatic breast cancer prediction (BG-MBC) that integrates graph information derived from the BERT model. In this model, nodes are constructed from patient medical records, while BERT embeddings are employed to vectorise representations of the words in histopathology reports, thereby capturing semantic information crucial for classification by employing three distinct approaches (namely univariate selection, extra trees classifier for feature importance, and Shapley values to identify the features that have the most significant impact). Identifying the most crucial 30 features out of 676 generated as embeddings during model training, our model further enhances its predictive capabilities. The BG-MBC model achieves outstanding accuracy, with a detection rate of 0.98 and an area under curve (AUC) of 0.98, in identifying MBC patients. This remarkable performance is credited to the model’s utilisation of attention scores generated by the LLM from histopathology reports, effectively capturing pertinent features for classification. Full article

The sneaker industry is continuing to expand at a fast rate and will be worth over USD 120 billion in the next few years. This is, in part due to social media and online retailers building hype around releases of limited-edition sneakers, which are usually collaborations between well-known global icons and footwear companies. These limited-edition sneakers are typically released in low quantities using an online raffle system, meaning only a few people can get their hands on them. As expected, this causes their value to skyrocket and has created an extremely lucrative resale market for sneakers. This has given rise to numerous counterfeit sneakers flooding the resale market, resulting in online platforms having to hand-verify a sneaker’s authenticity, which is an important but time-consuming procedure that slows the selling and buying process. To speed up the authentication process, Support Vector Machines and a convolutional neural network were used to classify images of fake and real sneakers and then their accuracies were compared to see which performed better. The results showed that the CNNs performed much better at this task than the SVMs with some accuracies over 95%. Therefore, a CNN is well equipped to be a sneaker authenticator and will be of great benefit to the reselling industry. Full article

Cardiovascular diseases present a significant global health challenge that emphasizes the critical need for developing accurate and more effective detection methods. Several studies have contributed valuable insights in this field, but it is still necessary to advance the predictive models and address the gaps in the existing detection approaches. For instance, some of the previous studies have not considered the challenge of imbalanced datasets, which can lead to biased predictions, especially when the datasets include minority classes. This study’s primary focus is the early detection of heart diseases, particularly myocardial infarction, using machine learning techniques. It tackles the challenge of imbalanced datasets by conducting a comprehensive literature review to identify effective strategies. Seven machine learning and deep learning classifiers, including K-Nearest Neighbors, Support Vector Machine, Logistic Regression, Convolutional Neural Network, Gradient Boost, XGBoost, and Random Forest, were deployed to enhance the accuracy of heart disease predictions. The research explores different classifiers and their performance, providing valuable insights for developing robust prediction models for myocardial infarction. The study’s outcomes emphasize the effectiveness of meticulously fine-tuning an XGBoost model for cardiovascular diseases. This optimization yields remarkable results: 98.50% accuracy, 99.14% precision, 98.29% recall, and a 98.71% F1 score. Such optimization significantly enhances the model’s diagnostic accuracy for heart disease. Full article

Ensuring consistent high water quality is paramount in water management planning. This paper addresses this objective by proposing an intelligent edge-cloud framework for water quality monitoring within the water distribution system (WDS). Various scenarios—cloud computing, edge computing, and hybrid edge-cloud computing—are applied to identify the most effective platform for the proposed framework. The first scenario brings the analysis closer to the data generation point (at the edge). The second and third scenarios combine both edge and cloud platforms for optimised performance. In the third scenario, sensor data are directly sent to the cloud for analysis. The proposed framework is rigorously tested across these scenarios. The results reveal that edge computing (scenario 1) outperforms cloud computing in terms of latency, throughput, and packet delivery ratio obtaining 20.33 ms, 148 Kb/s, and 97.47%, respectively. Notably, collaboration between the edge and cloud enhances the accuracy of classification models with an accuracy of up to 94.43%, this improvement was achieved while maintaining the energy consumption rate at the lowest value. In conclusion, our study demonstrates the effectiveness of the proposed intelligent edge-cloud framework in optimising water quality monitoring, and the superior performance of edge computing, coupled with collaborative edge-cloud strategies, underscores the practical viability of this approach. Full article

With the dramatic increase of the global population and with food insecurity increasing, it has become a major concern for both individuals and governments to fulfill the need for foods such as vegetables and fruits. Moreover, the desire for the consumption of healthy food, including fruit, has increased the need for applications in the field of agriculture that help to achieve better methods for fruit sorting and fruit disease prediction and classification. Automated fruit recognition is a potential solution to reduce the time and labor required to identify different fruits in situations such as retail stores during checkout, fruit processing centers during sorting, and orchards during harvest. Automating these processes reduces the need for human intervention, making them cheaper, faster, and immune to human error and biases. Past research in the field has focused mainly on the size, shape, and color features of fruits or employed convolutional neural networks (CNNs) for their classification. This study investigates the effectiveness of pre-trained deep learning models for fruit classification using two distinct datasets: Fruits-360 and the Fruit Recognition dataset. Four pre-trained models, DenseNet-201, Xception, MobileNetV3-Small, and ResNet-50, were chosen for the experiments based on their architecture and features. The results show that all models achieved almost 99% accuracy or higher with Fruits-360. With the Fruit Recognition dataset, DenseNet-201 and Xception achieved accuracies of around 98%. The good results exhibited by DenseNet-201 and Xception on both the datasets are remarkable, with DenseNet-201 attaining accuracies of 99.87% and 98.94%, and Xception attaining 99.13% and 97.73% accuracy, respectively, on Fruits-360 and the Fruit Recognition dataset. Full article

The wastewater quality index (WWQI) is one of the most significant methods of presenting meaningful values that reflect a fundamental characteristic of wastewater. Therefore, this study was performed to develop a prediction approach using WWQI for a regional wastewater treatment plant (WWTP) in Melaka, Malaysia. The regional system of WWTP provides a huge amount of registered data due to the many parameters recorded daily. A multivariate statistical analysis approach was applied to analyze the database. In this approach, principal component analysis (PCA) was used to reduce the dimensionality of datasets obtained from the field municipal WWTP, and multiple linear regression (MLR) was used to predict the performance of WWQI. Seven principal component analyses were derived where the eigenvalue was above 1.0, explaining 71.01% of the variance. A linear relationship was observed (R² = 0.85), p-value < 0.05, and residual values were uniformly distributed above and below the zero baselines. Therefore, the coefficients of the WWQI model are directly dependent on influent biological oxygen demand (BOD), effluent BOD, influent chemical oxygen demand (COD), and effluent COD values. The experimental results showed that the model performed well and can be used to predict WWQI for each WWTP individually and provide better achievements. Full article

The concept of molecular similarity has been commonly used in rational drug design, where structurally similar molecules are examined in molecular databases to retrieve functionally similar molecules. The most used conventional similarity methods used two-dimensional (2D) fingerprints to evaluate the similarity of molecules towards a target query. However, these descriptors include redundant and irrelevant features that might impact the performance of similarity searching methods. Thus, this study proposed a new approach for identifying the important features of molecules in chemical datasets based on the representation of the molecular features using Autoencoder (AE), with the aim of removing irrelevant and redundant features. The proposed approach experimented using the MDL Data Drug Report standard dataset (MDDR). Based on experimental findings, the proposed approach performed better than several existing benchmark similarity methods such as Tanimoto Similarity Method (TAN), Adapted Similarity Measure of Text Processing (ASMTP), and Quantum-Based Similarity Method (SQB). The results demonstrated that the performance achieved by the proposed approach has proven to be superior, particularly with the use of structurally heterogeneous datasets, where it yielded improved results compared to other previously used methods with the similar goal of improving molecular similarity searching. Full article

Machine learning has traditionally been solely performed on servers and high-performance machines. However, advances in chip technology have given us miniature libraries that fit in our pockets and mobile processors have vastly increased in capability narrowing the vast gap between the simple processors embedded in such things and their more complex cousins in personal computers. Thus, with the current advancement in these devices, in terms of processing power, energy storage and memory capacity, the opportunity has arisen to extract great value in having on-device machine learning for Internet of Things (IoT) devices. Implementing machine learning inference on edge devices has huge potential and is still in its early stages. However, it is already more powerful than most realise. In this paper, a step forward has been taken to understand the feasibility of running machine learning algorithms, both training and inference, on a Raspberry Pi, an embedded version of the Android operating system designed for IoT device development. Three different algorithms: Random Forests, Support Vector Machine (SVM) and Multi-Layer Perceptron, respectively, have been tested using ten diverse data sets on the Raspberry Pi to profile their performance in terms of speed (training and inference), accuracy, and power consumption. As a result of the conducted tests, the SVM algorithm proved to be slightly faster in inference and more efficient in power consumption, but the Random Forest algorithm exhibited the highest accuracy. In addition to the performance results, we will discuss their usability scenarios and the idea of implementing more complex and taxing algorithms such as Deep Learning on these small devices in more details. Full article