Repositorio Universidad Europea del Atlántico
Not a member yet
2719 research outputs found
Sort by
A deep learning approach to optimize remaining useful life prediction for Li-ion batteries
No full textAccurately predicting the remaining useful life (RUL) of lithium-ion (Li-ion) batteries is vital for improving battery performance and safety in applications such as consumer electronics and electric vehicles. While the prediction of RUL for these batteries is a well-established field, the current research refines RUL prediction methodologies by leveraging deep learning techniques, advancing prediction accuracy. This study proposes AccuCell Prodigy, a deep learning model that integrates auto-encoders and long short-term memory (LSTM) layers to enhance RUL prediction accuracy and efficiency. The model’s name reflects its precision (“AccuCell”) and predictive strength (“Prodigy”). The proposed methodology involves preparing a dataset of battery operational features, split using an 80–20 ratio for training and testing. Leveraging 22 variations of current (critical parameter) across three Li-ion cells, AccuCell Prodigy significantly reduces prediction errors, achieving a mean square error of 0.1305%, mean absolute error of 2.484%, and root mean square error of 3.613%, with a high R-squared value of 0.9849. These results highlight its robustness and potential for advancing battery health management
Enhanced interpretable thyroid disease diagnosis by leveraging synthetic oversampling and machine learning models
No full textThyroid illness encompasses a range of disorders affecting the thyroid gland, leading to either hyperthyroidism or hypothyroidism, which can significantly impact metabolism and overall health. Hypothyroidism can cause a slowdown in bodily processes, leading to symptoms such as fatigue, weight gain, depression, and cold sensitivity. Hyperthyroidism can lead to increased metabolism, causing symptoms like rapid weight loss, anxiety, irritability, and heart palpitations. Prompt diagnosis and appropriate treatment are crucial in managing thyroid disorders and improving patients’ quality of life. Thyroid illness affects millions worldwide and can significantly impact their quality of life if left untreated. This research aims to propose an effective artificial intelligence-based approach for the early diagnosis of thyroid illness. An open-access thyroid disease dataset based on 3,772 male and female patient observations is used for this research experiment. This study uses the nominal continuous synthetic minority oversampling technique (SMOTE-NC) for data balancing and a fine-tuned light gradient booster machine (LGBM) technique to diagnose thyroid illness and handle class imbalance problems. The proposed SNL (SMOTE-NC-LGBM) approach outperformed the state-of-the-art approach with high-accuracy performance scores of 0.96. We have also applied advanced machine learning and deep learning methods for comparison to evaluate performance. Hyperparameter optimizations are also conducted to enhance thyroid diagnosis performance. In addition, we have applied the explainable Artificial Intelligence (XAI) mechanism based on Shapley Additive exPlanations (SHAP) to enhance the transparency and interpretability of the proposed method by analyzing the decision-making processes. The proposed research revolutionizes the diagnosis of thyroid disorders efficiently and helps specialties overcome thyroid disorders early
Novel prehospital lactate cut-off estimation for mortality: a multicentre observational study
No full textObjectives: Point-of-care testing available in prehospital settings requires the establishment of new medical decision points. The aim of the present work was to determine the cut-off of the lactate threshold that activates alert triggers for all-cause 2-day mortality.
Design: Multicentre, prospective, ambulance-based, observational study.
Setting: Patients treated via emergency medical services (EMSs) and delivered to the emergency department between 2019 and 2023 were selected in Spain.
Participants: Adults with any acute disease.
Primary and secondary outcome measures: Epidemiological data, vital signs and prehospital point-of-care glucose and lactate levels were obtained. The outcome was all-cause 2-day in-hospital mortality. The cut-offs were obtained via three different methods: (i) indirect (which considers survivors and non-survivors), direct (which considers only survivors) assessment and lactate quartile. Additionally, the quartile approach was used to determine the differences in lactate distribution between survivors and non-survivors. Three different back-to-back studies with the same methodology were used.
Results: A total of 11 713 patients fulfilled the inclusion criteria. The mortality rate was 4.6% (542 patients). The difference in the median prehospital lactate concentration (mmol/L) between survivors and non-survivors was statistically significant (p<0.001): 2.29 (95% CI 1.43 to 3.38) and 7.14 (95% CI 5.11 to 9.71), respectively. Globally, the cut-off for all the studies combined was estimated by the direct method to be 3.71 mmol/L (95% CI 2.92 to 3.91), which was similar to the indirect value of 3.07 (95% CI 2.95 to 5.49) and the third quartile of 4.00. The mortality rate in patients who were less than 3.71 mmol/L was 0.004%, and that above that cut-off was 18%.
Conclusions: This study established a real-world lactate cut-off for 2-day in-hospital mortality of 3.71 mmol/L (95% CI 2.92 to 3.91) on the basis of data from the EMS. Considering this cut-off point could improve patient management via EMS services, allowing quick identification of patients at high risk of clinical worsening
Underwater Thermal Energy Harvesting: Frameworks, Challenges, Applications, and Future Investigation
No full textThis paper studies the latest and state-of-the-art underwater thermal energy harvesting algorithms and techniques designed in the latest decade (2014-2024). The techniques are classified based on their unique operations for energy harvesting. This classification includes thermal energy harvesting using a phase change material (PCM), thermoelectric generator (TEG) and multi-source harvesting. Every class of techniques is described by its operation using a schematic diagram and a mathematical model to fully understand its working principle. Moreover, every individual technique is also described in terms of its operation, amount of harvested energy/power and the aspect(s) where margin of further improvement exists. Also, a comparative analysis of the classified algorithms is performed with each other as well as with other underwater energy harvesting techniques (solar, piezoelectric, wave) to highlight their effectiveness and feasibility in a diverse set of underwater and various other applications. The classified techniques are also compared in terms of harvested output to indicate their harvesting efficiency. Furthermore, the publications made in the latest decade in terms of thermal energy harvesting using PCM, TEG and multi-source methods are also graphically depicted. Such a description of the studied techniques and classified methods is unique from the already existing underwater energy harvesting reviews in literature where an in-depth and thorough analysis is absent, rather only marginal description is given. The harvesting results indicate that hybrid (multi-source) and PCM methods have the greatest amount of harvested power and energy, respectively. Finally, the research challenges in underwater thermal energy harvesting are specified and areas of further research are highlighted for future investigation
In Vitro and In Vivo Insights into a Broccoli Byproduct as a Healthy Ingredient for the Management of Alzheimer’s Disease and Aging through Redox Biology
No full textBroccoli has gained popularity as a highly consumed vegetable due to its nutritional and health properties. This study aimed to evaluate the composition profile and the antioxidant capacity of a hydrophilic extract derived from broccoli byproducts, as well as its influence on redox biology, Alzheimer’s disease markers, and aging in the Caenorhabditis elegans model. The presence of glucosinolate was observed and antioxidant capacity was demonstrated both in vitro and in vivo. The in vitro acetylcholinesterase inhibitory capacity was quantified, and the treatment ameliorated the amyloid-β- and tau-induced proteotoxicity in transgenic strains via SOD-3 and SKN-1, respectively, and HSP-16.2 for both parameters. Furthermore, a preliminary study on aging indicated that the extract effectively reduced reactive oxygen species levels in aged worms and extended their lifespan. Utilizing broccoli byproducts for nutraceutical or functional foods could manage vegetable processing waste, enhancing productivity and sustainability while providing significant health benefits
Pneumonia Detection Using Chest Radiographs With Novel EfficientNetV2L Model
No full textPneumonia is a potentially life-threatening infectious disease that is typically diagnosed through physical examinations and diagnostic imaging techniques such as chest X-rays, ultrasounds or lung biopsies. Accurate diagnosis is crucial as wrong diagnosis, inadequate treatment or lack of treatment can cause serious consequences for patients and may become fatal. The advancements in deep learning have significantly contributed to aiding medical experts in diagnosing pneumonia by assisting in their decision-making process. By leveraging deep learning models, healthcare professionals can enhance diagnostic accuracy and make informed treatment decisions for patients suspected of having pneumonia. In this study, six deep learning models including CNN, InceptionResNetV2, Xception, VGG16, ResNet50 and EfficientNetV2L are implemented and evaluated. The study also incorporates the Adam optimizer, which effectively adjusts the epoch for all the models. The models are trained on a dataset of 5856 chest X-ray images and show 87.78%, 88.94%, 90.7%, 91.66%, 87.98% and 94.02% accuracy for CNN, InceptionResNetV2, Xception, VGG16, ResNet50 and EfficientNetV2L, respectively. Notably, EfficientNetV2L demonstrates the highest accuracy and proves its robustness for pneumonia detection. These findings highlight the potential of deep learning models in accurately detecting and predicting pneumonia based on chest X-ray images, providing valuable support in clinical decision-making and improving patient treatment
Bridging personality dimensions and eating symptoms: A transdiagnostic network approach
No full textObjective
Eating disorders (ED) have recently been studied from a network approach, conceptualising them as a complex system of interconnected variables, while highlighting the role of non-ED symptoms and personality dimensions. This study aims to explore the connections between personality and ED symptoms, identify central nodes, and compare the EDs network to a healthy control network.
Methods
We employed network analysis to examine the personality-ED symptom connections in 329 individuals with an ED diagnosis and 192 healthy controls. We estimated a regularised partial correlation network and the indices of centrality and bridge centrality to identify the most influential nodes for each group. Network differences between groups were also examined.
Results
Low Self-Directedness and high Harm avoidance emerged as central bridge nodes, displaying the strongest relationship with ED symptoms. Both networks differed in their global connectivity and structure, although no differences were found in bridge centrality and centrality indices.
Conclusions
These findings shed light on the role of personality dimensions, such as Self-Directedness and Harm Avoidance in the maintenance of ED psychopathology, supporting the transdiagnostic conceptualisation of ED. This study advances a deeper understanding of the complex interplay between personality dimensions and ED symptoms, offering potential directions for clinical interventions
Remote Sensing and Environmental Monitoring Analysis of Pigment Migrations in Cave of Altamira’s Prehistoric Paintings
No full textThe conservation of Cultural Heritage in cave environments, especially those hosting cave art, requires comprehensive conservation strategies to mitigate degradation risks derived from climatic influences and human activities. This study, focused on the Polychrome Hall of the Cave of Altamira, highlights the importance of integrating remote sensing methodologies to carry out effective conservation actions. By coupling a georeferenced Ground Penetrating Radar (GPR) with a 1.6 GHz central-frequency antenna along with photogrammetry, we conducted non-invasive and high-resolution 3D studies to map preferential moisture pathways from the surface of the ceiling to the first 50 cm internally of the limestone structure. In parallel, we monitored the dynamics of surface water on the Ceiling and its correlation with pigment and other substance migrations. By standardizing our methodology, we aim to increase knowledge about the dynamics of infiltration water, which will enhance our understanding of the deterioration processes affecting cave paintings related to infiltration water. This will enable us to improve conservation strategies, suggesting possible indirect measures to reverse active deterioration processes. Integrating remote sensing techniques with geospatial analysis will aid in the validation and calibration of collected data, allowing for stronger interpretations of subsurface structures and conditions. All of this puts us in a position to contribute to the development of effective conservation methodologies, reduce alteration risks, and promote sustainable development practices, thus emphasizing the importance of remote sensing in safeguarding Cultural Heritage
Natural Language Processing-Based Software Testing: A Systematic Literature Review
No full textNew approaches to software testing are required due to the rising complexity of today’s software applications and the rapid growth of software engineering practices. Among these methods, one that has shown promise is the introduction of Natural Language Processing (NLP) tools to software testing practices. NLP has witnessed a rise in popularity within all IT fields, especially in software engineering, where its use has improved the way we extract information from textual data. The goal of this systematic literature review (SLR) is to provide an in-depth analysis of the present body of the literature on the expanding subject of NLP-based software testing. Through a repeatable process, that takes into account the quality of the research, we examined 24 papers extracted from Web of Science and Scopus databases to extract insights about the usage of NLP techniques in the field of software testing. Requirements analysis and test case generation popped up as the most hot topics in the field. We also explored NLP techniques, software testing types, machine/deep learning algorithms, and NLP tools and frameworks used in the studied body of literature. This study also stressed some recurrent open challenges that need further work in future research such as the generalization of the NLP algorithm across domains and languages and the ambiguity in the natural language requirements. Software testing professionals and researchers can get important insights from the findings of this SLR, which will help them comprehend the advantages and challenges of using NLP in software testing
Nutritional, functional, and safety characterization of the edible larva of the South American palm weevil (chontacuro) Rhynchophorus palmarum L. from Amazonian Ecuador
No full textEdible insects represent a viable option to address the current need for nutritious, safe, and eco-friendly foods. People native to the Amazon region have a long-standing tradition of consuming edible insects that are relatively unknown elsewhere. This research aimed to characterize the chemical, nutritional, and microbiological composition of the edible larva of the palm weevil Rhynchophorus palmarum L. (chontacuro) from the Amazonian lowlands of Ecuador. The larvae proved to be rich in lipids (∼50 %), proteins (∼20 %), fiber (∼6 %), and oleic acid, one of their predominant fatty acids along with palmitic acid. The larvae are also rich in vitamins (B6, B9, A, and E) and are a source of β-carotene, calcium, potassium, magnesium, and phosphorus. No evidence of toxic elements (metals) or pathogenic microorganisms was observed. In general, chontacuro larvae proved to be a safe and nutritious food, managing to fully or partially cover several of the Dietary Reference Intakes for several nutrients