Ayala-Chauvin, Manuel Ignacio
Preferred name
Ayala-Chauvin, Manuel Ignacio
Main Affiliation
Ambato
Email
mayala@uti.edu.ec
ORCID
0000-0002-3911-377X
Scopus Author ID
57215428914
43 results
Now showing 1 - 10 of 43
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Road Drainage Infrastructure Diagnostics and Deficiency Indexing in ENSO-Vulnerable Andean Corridors: A STEM–PjBL Field Assessment(MDPI AG, 2026-05-15); Benavides-Ortega, Leirys MaríaRoad drainage infrastructure in ENSO-vulnerable Andean regions faces compounding threats from climatic variability, geometric inadequacy, and systemic maintenance neglect. This study presents a STEM-integrated Project-Based Learning (PjBL) diagnostic framework applied to 42 road segments along corridors connecting Loja, Ecuador, selected through a purposive-stratified spatial-coverage protocol. Using ArcGIS Survey123, standardised field data were collected on structure presence, geometry, failure modes, and condition across four structure types: crown gutters, road gutters, hydraulic chutes, and culverts. The Composite Drainage Deficiency Index (DDI, 0–100) was derived from five equally weighted binary indicators and validated through Monte Carlo Dirichlet weight-perturbation analysis and jackknife leave-one-out resampling, confirming rank-order invariance to admissible alternative weightings. The results reveal severe systemic deficiencies, including crown gutters absent at 88.1% (95% CI: 75.0–94.8) and road gutters at 81.0% (95% CI: 66.7–90.0) of sites. Every segment exhibited at least one drainage failure (100%; 95% CI: 91.6–100). The DDI identified 73.8% of segments in the High or Critical band (DDI ≥ 60; mean = 60.2 ± 20.4). Hierarchical clustering isolated one geometric outlier whose exclusion altered the aggregate metrics by <1.2%. These findings establish a georeferenced baseline for maintenance prioritisation and validate the methodological reproducibility of academically integrated field protocols for infrastructure diagnostics. © 2026 by the authors. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, U-Net Models for Breast Cancer Detection: Improving Diagnostic Accuracy and Specificity(2024); Elena Blanco-RomeroBreast cancer remains a critical global health issue, necessitating continuous research and innovative approaches for diagnosis, treatment, and prevention. This study evaluates the effectiveness of U -Net models in enhancing diagnostic precision and efficiency using real hospital samples. We aim to improve key diagnostic metrics such as accuracy, sensitivity, and specificity through the application of U-Net models. Our image classification model, tailored for 256 × 256 × 3 input images, excels in detecting and categorizing tumor cells. The architecture begins with initial convolutional layers featuring 64 filters, progresses to layers with 128 filters, and includes a Dropout layer to prevent overfitting. The deep network for object detection utilizes both region proposal and regression/classification approaches, achieving 92.27% confidence and 100% accuracy. Additionally, our deep learning algorithms accurately segment nuclei in histopathological images, employing a clustering strategy that delivers 88.81% confidence and 100% accuracy. Visual results demonstrate precise tumor cell localization and prediction confidence. Performance metrics from ten experimental runs indicate average confidence levels between 74.19% and 92.31%, with 90.0% accuracy and specificity in benign analysis. The model's ability to classify non-carcinomas versus carcinomas achieved an AUC of 0.78, illustrating its effective differentiation between classes.10 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Predicting Academic Performance in Mathematics Using Machine Learning AlgorithmsSeveral factors, directly and indirectly, influence students’ performance in their various activities. Children and adolescents in the education process generate enormous data that could be analyzed to promote changes in current educational models. Therefore, this study proposes using machine learning algorithms to evaluate the variables influencing mathematics achievement. Three models were developed to identify behavioral patterns such as passing or failing achievement. On the one hand, numerical variables such as grades in exams of other subjects or entrance to higher education and categorical variables such as institution financing, student’s ethnicity, and gender, among others, are analyzed. The methodology applied was based on CRISP-DM, starting with the debugging of the database with the support of the Python library, Sklearn. The algorithms used are Decision Tree (DT), Naive Bayes (NB), and Random Forest (RF), the last one being the best, with 92% accuracy, 98% recall, and 97% recovery. As mentioned above, the attributes that best contribute to the model are the entrance exam score for higher education, grade exam, and achievement scores in linguistic, scientific, and social studies domains. This confirms the existence of data that help to develop models that can be used to improve curricula and regional education regulations. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.39 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Performance and Real-World Variability of Predictive Maintenance Models for Vehicle Fleets(2024); ; Elena Blanco-RomeroThis study presents a comprehensive evaluation of predictive maintenance models for vehicle fleets, detailing a sequence of systematic steps to ensure model performance and address real-world variability. The process begins with database creation and data preprocessing, where relevant maintenance records are filtered, and datetime columns are converted to facilitate time-based calculations. Grouping and aggregation techniques are then applied to count occurrences of specific maintenance activities and identify common failure types. For model training, we define a neural network architecture comprising dense and dropout layers to mitigate overfitting, compile the model with suitable loss functions and optimizers, and train it using the prepared data. The trained model, along with the scaler and encoder, is saved for future use. To augment the dataset, synthetic data is generated using the Faker library and random distributions, with added noise to mimic real-world variability. Preprocessing steps are reapplied to this synthetic data to ensure consistency. By implementing this neural network, we achieved a sensitivity of 0.93 and an ROC of 0.71. Following these detailed steps, we develop a robust predictive maintenance model that effectively identifies failures and non-failures, ultimately enhancing the reliability and efficiency of vehicle fleet management.27 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Evaluation of Accessibility on the PAR Platform from the Perspective of Physicians(2024)23 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Augmented Reality Application with Multimedia Content to Support Primary Education(2023); ; 31 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Anthropization and growth of the electricity grid as variables for the analysis of urban infrastructure(2020); ; 14 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Prevention of Failures in the Footwear Production Process by Applying Machine Learning(2022); 25 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Design of a Gravity Ropeway in Nepal: A Methodological Analysis for Appropriate Technologies(2024)18 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Sustainable Development in Higher Education Curricula for Software Engineering Chairs(2023); ; 41