Learning Approach Research Articles

A multiscale approach for biomedical machine learning

A multiscale approach for biomedical machine learning

Linear Parameter Varying and Reinforcement Learning Approaches for Trajectory Tracking Controller of Autonomous Vehicles

This research focuses on controlling the motion trajectory of autonomous vehicles by using a combination of two high-performance control methods: Linear Parameter Varying (LPV) and Reinforcement Learning (RL). First, a single-track motion model is researched and developed with coordinate systems to determine the car's motion trajectory through signals from GPS. Then, the LPV control method is used to design a controller to control the car's motion trajectory. Reinforcement learning method with detailed training procedures is used to combine with the advantages of LPV controller. Finally, the simulation results are evaluated in the time domain through the use of specialized CarSim software, which clearly demonstrates the superiority of the research method.

Detecting rumors in social media using emotion based deep learning approach

Detecting rumors in social media using emotion based deep learning approach

Hybrid Deep Learning Approach with Feature Engineering for Enhanced Pulmonary Nodule Diagnosis

Hybrid Deep Learning Approach with Feature Engineering for Enhanced Pulmonary Nodule Diagnosis

Enhancing the Minimum Awareness Failure Distance in V2X Communications: A Deep Reinforcement Learning Approach

Enhancing the Minimum Awareness Failure Distance in V2X Communications: A Deep Reinforcement Learning Approach

Correction to: Deep learning approaches to detect breast cancer: a comprehensive review

Correction to: Deep learning approaches to detect breast cancer: a comprehensive review

A Machine Learning Approach for Predicting Biochemical Outcome After PSMA-PET–Guided Salvage Radiotherapy in Recurrent Prostate Cancer After Radical Prostatectomy: Retrospective Study

A Machine Learning Approach for Predicting Biochemical Outcome After PSMA-PET–Guided Salvage Radiotherapy in Recurrent Prostate Cancer After Radical Prostatectomy: Retrospective Study

Prediction of preterm birth in multiparous women using logistic regression and machine learning approaches

To predict preterm birth (PTB) in multiparous women, comparing machine learning approaches with traditional logistic regression. A population-based cohort study was conducted using data from the Ontario Better Outcomes Registry and Network (BORN). The cohort included all multiparous women who delivered a singleton birth at 20–42 weeks’ gestation in an Ontario hospital between April 1, 2012 and March 31, 2014. The primary outcome was PTB < 37 weeks, with spontaneous PTB analyzed as a secondary outcome. Stepwise logistic regression and the Boruta machine learning were used to select the important variables during the first and second trimester. For building prediction models, the whole data set were divided for the two independent parts: two-third for training the classifiers (Logistic regression, random forests, decision trees, and artificial neural networks) and one-third for model validation. Then, the training data set were balanced by random over sampling technique. The best hyper parameters were obtained by the tenfold cross validation. The performance of all models was evaluated by sensitivity, specificity, positive predictive value, negative predictive value, and the area under the receiver operating characteristics (AUC). The cohort included 145,846 births, of which 8125 (5.57%) were preterm. In first-trimester models, the strongest predictors of PTB were previous PTB, preexisting diabetes, and abnormal pregnancy‐associated plasma protein-A. In the testing data set, the highest predictive ability was seen for artificial neural networks, with an area under the receiver operating characteristic curve (AUC) of 68.8% (95% CI 67.6–70.1%). In second-trimester models, addition of infant sex, attendance at first-trimester appointment, medication exposure, and abnormal alpha-fetoprotein concentrations increased the AUC to 72.1% (95% CI 71.1–73.1%) with logistic regression. With the inclusion of the variable complications during pregnancy, the AUC increased to 80.5% (95% CI 79.6–81.5%) using logistic regression. For both overall and spontaneous PTB, during both the first and second trimesters, models yielded negative predictive values of 97%. Overall, machine learning and logistic regression produced similar performance for prediction of PTB. For overall and spontaneous PTB, both first- and second-trimester models provided negative predictive values of ~ 97%, higher than that of fetal fibronectin.

Calibrating machine learning approaches for probability estimation: A short expansion.

Calibrating machine learning approaches for probability estimation: A short expansion.

Enhanced Osteoporosis Detection Using Artificial Intelligence: A Deep Learning Approach to Panoramic Radiographs with an Emphasis on the Mental Foramen

Osteoporosis, a skeletal disorder, is expected to affect 60% of women aged over 50 years. Dual-energy X-ray absorptiometry (DXA) scans, the current gold standard, are typically used post-fracture, highlighting the need for early detection tools. Panoramic radiographs (PRs), common in annual dental evaluations, have been explored for osteoporosis detection using deep learning, but methodological flaws have cast doubt on otherwise optimistic results. This study aims to develop a robust artificial intelligence (AI) application for accurate osteoporosis identification in PRs, contributing to early and reliable diagnostics. A total of 250 PRs from three groups (A: osteoporosis group, B: non-osteoporosis group matching A in age and gender, C: non-osteoporosis group differing from A in age and gender) were cropped to the mental foramen region. A pretrained convolutional neural network (CNN) classifier was used for training, testing, and validation with a random split of the dataset into subsets (A vs. B, A vs. C). Detection accuracy and area under the curve (AUC) were calculated. The method achieved an F1 score of 0.74 and an AUC of 0.8401 (A vs. B). For young patients (A vs. C), it performed with 98% accuracy and an AUC of 0.9812. This study presents a proof-of-concept algorithm, demonstrating the potential of deep learning to identify osteoporosis in dental radiographs. It also highlights the importance of methodological rigor, as not all optimistic results are credible.

Multiview EEG signal analysis for diagnosis of schizophrenia: an optimized deep learning approach

Multiview EEG signal analysis for diagnosis of schizophrenia: an optimized deep learning approach

A Robust and Rapid Grid-Based Machine Learning Approach for Inside and Off-Network Earthquakes Classification in Dynamically Changing Seismic Networks

A Robust and Rapid Grid-Based Machine Learning Approach for Inside and Off-Network Earthquakes Classification in Dynamically Changing Seismic Networks

Diabetic retinopathy detection and categorizing using a lightweight deep learning approach

Diabetic retinopathy is an ocular disorder that has the potential to result in visual impairment and complete loss of vision in those diagnosed with diabetes. This illness affects the retinal blood vessels inside the light-sensitive tissue layer at the posterior of the eye, known as the retina. This paper presents a complete approach to diagnosing and categorizing diabetic retinopathy using deep learning models. A lightweight Convolutional Neural Network (CNN) is used to detect diabetic retinopathy in fundus images. This CNN has been developed to have fewer parameters and calculations, making it suited for resource-constrained environments while retaining decent performance. The categorization of diabetic retinopathy is carried out with the help of EfficientNet. This model uses an innovative compound scaling approach to strike a balance between the model's depth, width, and resolution. As a result, it maximizes computing efficiency while preserving high accuracy. The proposed detection model obtained an accuracy of 95%, and the classification model produced an accuracy of 84%.

Surface and deep learning: a blended learning approach in preclinical years of medical school

BackgroundSignificant challenges are arising around how to best enable peer communities, broaden educational reach, and innovate in pedagogy. While digital education can address these challenges, digital elements alone do not guarantee effective learning. This study reports a blended learning approach integrating online and face-to-face components, guided by the Student Approaches to Learning framework.MethodsThis study was carried out investigating learning in first and second year medical students over two academic years, 2019/20 and 2020/21. We evaluated: (1) comparison of students engaged with blended learning and traditional learning; and (2) student learning engaged with blended learning approach over a two-year preclinical curriculum. A revised two-factor study process questionnaire (R-SPQ-2F) evaluated students’ surface/deep learning before and after an academic year. Learning experience (LE) questionnaire was administered over the domains of learning engagement, and outcomes of learning approach. In-depth interviews were carried out to understand the context of students’ responses to the R-SPQ-2F and LE questionnaires.ResultsThe R-SPQ-2F analysis indicated first year students maintained deep learning but second year students became neutral across the academic year, regardless of learning approach, with workload contributing to this outcome. R-SPQ-2F sub-scales showed that students engaged with blended learning maintained an intrinsic interest to learning, as compared to traditional learning which led to surface learning motives. The LE questionnaire showed students engaged with blended learning had deeper subject interest, and more positive perceptions of workload, feedback, and effectively developed skills and knowledge. However, peer interactions from blended learning were significantly lacking. In-depth interviews revealed that the flexibility and multi-modality of blended learning enabled learning, but the best use of these features require teacher support. Online interactions could be cultivated through intentional institutional efforts.ConclusionsThis study highlights the importance of designing blended learning that leverages technology-enabled flexibility while prioritising collaborative, learner-centred spaces for deep engagement and knowledge construction.

Detection and diagnosis of diabetic eye diseases using two phase transfer learning approach

Background Early diagnosis and treatment of diabetic eye disease (DED) improve prognosis and lessen the possibility of permanent vision loss. Screening of retinal fundus images is a significant process widely employed for diagnosing patients with DED or other eye problems. However, considerable time and effort are required to detect these images manually. Methods Deep learning approaches in machine learning have attained superior performance for the binary classification of healthy and pathological retinal fundus images. In contrast, multi-class retinal eye disease classification is still a difficult task. Therefore, a two-phase transfer learning approach is developed in this research for automated classification and segmentation of multi-class DED pathologies. Results In the first step, a Modified ResNet-50 model pre-trained on the ImageNet dataset was transferred and learned to classify normal diabetic macular edema (DME), diabetic retinopathy, glaucoma, and cataracts. In the second step, the defective region of multiple eye diseases is segmented using the transfer learning-based DenseUNet model. From the publicly accessible dataset, the suggested model is assessed using several retinal fundus images. Our proposed model for multi-class classification achieves a maximum specificity of 99.73%, a sensitivity of 99.54%, and an accuracy of 99.67%.

A variant-informed decision support system for tackling COVID-19: a transfer learning and multi-attribute decision-making approach

The global impact of the COVID-19 pandemic, characterized by its extensive societal, economic, and environmental challenges, escalated with the emergence of variants of concern (VOCs) in 2020. Governments, grappling with the unpredictable evolution of VOCs, faced the need for agile decision support systems to safeguard nations effectively. This article introduces the Variant-Informed Decision Support System (VIDSS), designed to dynamically adapt to each variant of concern’s unique characteristics. Utilizing multi-attribute decision-making (MADM) techniques, VIDSS assesses a country’s performance by considering improvements relative to its past state and comparing it with others. The study incorporates transfer learning, leveraging insights from forecast models of previous VOCs to enhance predictions for future variants. This proactive approach harnesses historical data, contributing to more accurate forecasting amid evolving COVID-19 challenges. Results reveal that the VIDSS framework, through rigorous K-fold cross-validation, achieves robust predictive accuracy, with neural network models significantly benefiting from transfer learning. The proposed hybrid MADM approach integrated approaches yield insightful scores for each country, highlighting positive and negative criteria influencing COVID-19 spread. Additionally, feature importance, illustrated through SHAP plots, varies across variants, underscoring the evolving nature of the pandemic. Notably, vaccination rates, intensive care unit (ICU) patient numbers, and weekly hospital admissions consistently emerge as critical features, guiding effective pandemic responses. These findings demonstrate that leveraging past VOC data significantly improves future variant predictions, offering valuable insights for policymakers to optimize strategies and allocate resources effectively. VIDSS thus stands as a pivotal tool in navigating the complexities of COVID-19, providing dynamic, data-driven decision support in a continually evolving landscape.

Enhancing Fake News Detection with Word Embedding: A Machine Learning and Deep Learning Approach

The widespread dissemination of fake news on social media has necessitated the development of more sophisticated detection methods to maintain information integrity. This research systematically investigates the effectiveness of different word embedding techniques—TF-IDF, Word2Vec, and FastText—when applied to a variety of machine learning (ML) and deep learning (DL) models for fake news detection. Leveraging the TruthSeeker dataset, which includes a diverse set of labeled news articles and social media posts spanning over a decade, we evaluated the performance of classifiers such as Support Vector Machines (SVMs), Multilayer Perceptrons (MLPs), and Convolutional Neural Networks (CNNs). Our analysis demonstrates that SVMs using TF-IDF embeddings and CNNs employing TF-IDF embeddings achieve the highest overall performance in terms of accuracy, precision, recall, and F1 score. These results suggest that TF-IDF, with its capacity to highlight discriminative features in text, enhances the performance of models like SVMs, which are adept at handling sparse data representations. Additionally, CNNs benefit from TF-IDF by effectively capturing localized features and patterns within the textual data. In contrast, while Word2Vec and FastText embeddings capture semantic and syntactic nuances, they introduce complexities that may not always benefit traditional ML models like MLPs or SVMs, which could explain their relatively lower performance in some cases. This study emphasizes the importance of selecting appropriate embedding techniques based on the model architecture to maximize fake news detection performance. Future research should consider integrating contextual embeddings and exploring hybrid model architectures to further enhance detection capabilities. These findings contribute to the ongoing development of advanced computational tools for combating misinformation.

Estimating the parameters of epidemic spread on two-layer random graphs: a classical and a neural network approach

In this paper, we propose and analyse various methods for estimating the infection parameter τ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ au $$\\end{document} in an SIR (susceptible-infected-recovered) epidemic spread model simulated on two-layer random graphs with a preferential attachment component. The statistics that can be used for the estimates is the number of susceptible and infected vertices. Our classical approach is based on the maximum likelihood method, while the machine learning approach uses a graph neural network (GNN). The underlying graph has a layer consisting of disjoint cliques (representing e.g. households) and a random second layer with a preferential attachment structure. Our simulation study reveals that the estimation is poorer at the beginning of the epidemic, for larger preferential attachment parameter of the graph, and for larger τ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ au $$\\end{document}. As for the neural networks, we find that dense networks offer better training datasets. Moreover, GNN perfomance is measured better using the l2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$l_2$$\\end{document} loss function rather than cross-entropy.

Grid Search Optimized Machine Learning based Modeling of CO2 Emissions Prediction from Cars for Sustainable Environment

Carbon emissions have increased dramatically because of industrialization, trapping heat in the atmosphere and hastening climate change. This is a serious threat to the wealth, security, and well-being of the world. The effects are extensive, ranging from severe weather, disease outbreaks, and economic disruption to food insecurity and water scarcity. The World Health Organization (WHO) has determined that climate change poses the greatest threat to public health in the twenty-first century. Thus, precise CO2 emissions have emerged as a crucial concern in recent times. Several studies have tried to forecast the amount CO2 from industry and power plant using statistical analysis. Efficiency, robustness and diverse application was the limitation of the study. In this study, we have proposed an AI based model that is able to predict the amounts of CO2 emissions from cars. We applied a grid search-optimized machine learning approach using the publicly available Canadian dataset. Incorporation of different statistical analyses and preprocessing techniques such as duplicate data management, outlier rejection, scaling contributed to enhance the quality of the dataset. Later, grid search techniques were applied to tune the KNN, RF, and SVR models. The approach has enhanced the performance of CO2 emissions prediction. In the study, we further used the explainability of the random forest model to check the bias and fairness of predictability. MSE, RMSE, and R-squared metrics of the proposed approach were the highest as the state of the art.

Experimental investigation on sustainable machining of Monel using vegetable oils as cutting fluids and machine learning-based surface roughness prediction

Abstract The material processing industry is anticipated to mitigate environmental degradation. The protocols established by the International Organisation for Standardisation were adhered to. As a result, it would be prudent to investigate the feasibility of removing cutting fluids from the machining process. This study discusses an environmentally-friendly machining technique for turning nickel-based alloy Monel-500, which evaluates four different cooling conditions: dry machining, flood machining, Co-MQL (coconut oil), and Rb-MQL (Rice Bran Oil). These conditions were tested by experimenting with various machining parameters to investigate four aspects of the turning process: surface finish,cutting temperature, tool wear and chip morphology. Rice bran oil is considered eco-friendly compared to synthetic cutting fluids, and employing it in minimum quantity is economical and helps improve the machined workpiece's surface finish. The investigation has been further extended by applying machine learning algorithms to predict surface roughness, utilising two logical regressions implemented in Python. Among the two machine learning approaches, the random forest regression technique has demonstrated superior results, achieving a prediction accuracy of 99.8%. Consequently, a decision tree has been developed using this regression model to predict the surface roughness. The structured analysis of the decision tree provides more accurate conclusions, offering flexibility in adjusting parameters and expanding options for operation. As a result, the decision tree approach enables the efficient utilisation of production resources and enhances production capacity by making informed choices about cooling methods during the turning process. Rb-MQL has performed better in all aspects than the other three cooling conditions. When comparing machining under dry conditions, flood cooling, Co-MQL, and Rb-MQL (rice bran oil) reduce the tooltip temperature by 39.5%,25.45 and 24.11 %, respectively. Rb-MQL reduced surface roughness by 28.23%,43.59 and 60.49 % in contrast with machining under dry, flood, and Co-MQL.&amp;#xD;