Ayala-Chauvin, Manuel Ignacio

Preferred name

Main Affiliation

Centro de Investigación de Ciencias Humanas y de la Educación

Ambato

mayala@uti.edu.ec

ORCID

0000-0002-3911-377X

Scopus Author ID

57215428914

Now showing 1 - 10 of 42

Some of the metrics are blocked by your
consent settings
U-Net Models for Breast Cancer Detection: Improving Diagnostic Accuracy and Specificity
(2024)
Dayanara Yánez-Arcos
;
Ayala-Chauvin, Manuel Ignacio
;
Elena Blanco-Romero
Breast 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 your
consent settings
IoT Monitoring to Control a Bicycle Parking Lot
(2022)
Ruales, María Belén
;
Lara-Alvarez P.
;
Riba C.
;
Ayala-Chauvin, Manuel Ignacio
In recent years, the development of new technologies has improved the management of resources and services at the urban level. In this sense, several cities worldwide have developed intelligent infrastructures such as Smart Cities in which, through data collection and management, they aim to achieve social, environmental and economic improvements. Innovative bike racks are a promising solution to traffic-related problems in major cities around the world; however, there is a lack of low-cost solutions for controlling and monitoring bike racks and thus boosting the mobility of cyclists. This paper presents a system to monitor and control a bicycle parking lot. In order to achieve this goal, software and hardware specifications were defined and characterised by the control system. The conceptual design and detail of the prototype and the materialisation proceeded, where technology with ESP8266 microcontrollers and Raspberry Pi+Ethernet/WiFi microprocessors was used in the MQTT communication protocol to implement its architecture. The system implements in the bicycle parking lot of the Universidad Tecnológica Indoamérica. The series of data collected allowed for determining the frequency of use. With this, a database creates where the information on the frequency of use of bicycles is stored. Finally, through a mobile application, the availability of parking spaces can be consulted, and bikes in the parking lot can monitor. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
43
Some of the metrics are blocked by your
consent settings
Predicting Academic Performance in Mathematics Using Machine Learning Algorithms
(2022)
Espinosa Pinos, Carlos Alberto
;
Ayala-Chauvin, Manuel Ignacio
;
Buele, Jorge
Several 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 your
consent settings
Data Analysis for Performance Improvement of University Students Using IoT
(2024)
Ayala-Chauvin, Manuel Ignacio
;
Lara-Álvarez P.
;
Castro R.
Using Internet of Things (IoT) devices and data analysis techniques can potentially transform how universities approach improving student achievement. In this sense, the project is based on implementing a remotely operated pneumatic bank applying IoT for university education. With this technology, it is possible to obtain information about the factors that impact student achievement and design targeted interventions to help students improve their performance. The control system with low-cost technology was developed with Raspberry Pi, AnyDesk, and Canvas LMS for the remote connection. The experiment was carried out with two groups of 7 people, and it was identified that there are correlations of 0.87 and 0.62 between the performance of the students and the time they dedicate to studying and the hours they spend on the platform; this suggests a positive correlation between these variables. Therefore, as students spend more time studying and spending more hours on the platform, they are more likely to achieve better academic results. On the other hand, the study time of the group of students who used the bank remotely increased by 32% compared to those who used the bank in person; therefore, we can infer that with the implementation of the IoT, the use of the system is encouraged. Finally, based on the insights gained from the analysis, targeted interventions can be designed to help students improve their academic performance. © 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.
36
Some of the metrics are blocked by your
consent settings
Power Flow Optimization in Electrical Networks using Gekko
(2025)
Ayala-Chauvin, Manuel Ignacio
;
Avilés-Castillo, Fátima
;
Riba-Romeva, Carles
Power flow optimization in the electrical grid is critical to improve the stability and performance of power systems. The main challenge lies in finding an optimal distribution of power generation that meets the constraints imposed by the grid, such as voltage limits and power system stability conditions. The objective of this research was to evaluate the performance of Gekko in power flow optimization in electrical grids. To do so, a comparison was made with SciPy, a widely used benchmark framework in numerical optimization in order to assess their relative efficiency in problems with complex constraints. The comparison is based on metrics such as solution accuracy, convergence speed, and number of objective function evaluations. The results showed that both methods achieved the same objective value: SciPy (19.7) and Gekko (19.7). However, SciPy was slightly faster (0.01496 seconds vs. 0.0191 seconds), but required 60 objective function evaluations. In contrast, Gekko demonstrated greater computational efficiency, reducing the number of evaluations required for convergence. While SciPy is more efficient on small problems with explicit constraints, Gekko offers greater flexibility on problems with more complex constraints, making it more suitable for larger power systems.
37
Some of the metrics are blocked by your
consent settings
Performance and Real-World Variability of Predictive Maintenance Models for Vehicle Fleets
(2024)
Avilés-Castillo, Fátima
;
Dayanara Yánez-Arcos
;
Ayala-Chauvin, Manuel Ignacio
;
Elena Blanco-Romero
This 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 your
consent settings
Evaluation of Accessibility on the PAR Platform from the Perspective of Physicians
(2024)
Patricia Acosta-Vargas
;
Gloria Acosta-Vargas
;
Marco Santórum
;
Mayra Carrión-Toro
;
Ayala-Chauvin, Manuel Ignacio
23
Some of the metrics are blocked by your
consent settings
Augmented Reality Application with Multimedia Content to Support Primary Education
(2023)
Buele, Jorge
;
Espinoza J.
;
Ruales, María Belén
;
Camino-Morejón V.M.
;
Ayala-Chauvin, Manuel Ignacio
Education is continually reinventing itself to meet the growing needs of learners. In the context of an emergency, confinement and mobilization difficulties have necessitated the adoption of online education. This new offer brings technological means to conventional processes, but its main disadvantage is the limited access in some places. This has motivated the present study, which proposes the development of a mobile application using augmented reality (AR) to complement primary education. This application aims to disseminate multimedia content without needing a stable internet connection. A high-performance computer is required for the design and a mid-range smartphone for its execution. Four scripts are generated in c# programming language using Visual Studio. The environment and the three-dimensional objects are developed using Unity software and the packages ARFoundation, Unity MARS, and Vuforia. Two groups of 13 children each participated in the experimental testing. The experimental group used the application for two weeks to complement the virtual classes in the subjects of language, natural sciences, and mathematics. The analysis with SPSS software shows a statistically significant increase in the average grades compared to the control group. This research shows that the use of technology can contribute to improving the current teaching-learning processes. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
31
Some of the metrics are blocked by your
consent settings
Anthropization and growth of the electricity grid as variables for the analysis of urban infrastructure
(2020)
Ayala-Chauvin, Manuel Ignacio
;
Huaraca D.
;
Varela Aldas, José
;
Ordóñez A.
;
Riba G.
City growth goes together with the development of infrastructure, and the power network is one of the most relevant towards economic development. The study of urban infrastructure through the analysis of anthropization coupled with power network growth can produce a tool that supports sustainable infrastructure planning, both economic and environmental. The case study focuses on Ambato, Ecuador, in the period from 1950 to 2019, and assesses quantitatively the changes in the city layout and the evolution of its power network. The data are adjusted to a sigmoid-type objective function through a non-linear least squares problem, that is solved using the Broyden-Fletcher-Goldfarb-Shanno (BFGS) method. Anthropization data show how the urban area grew during the study period: 37% (1950-1960), 53% (1960-1970), 80% (1970-1980), 35% (1980-1990), 39% (1990-2000), 38% (2000-2010), and 11% (2010-2019), mostly at the expense of agricultural land. The forecast for new power network users by 2050 yields a result of 203,630 total users with a population density of 4850 people/km2. The conclusion is that this type of analysis can help city planners and decision makers further understand city and infrastructure growth dynamics and produce policies that bolster sustainable city growth. © 2020 by the authors.
14
Some of the metrics are blocked by your
consent settings
Prevention of Failures in the Footwear Production Process by Applying Machine Learning
(2022)
Tierra-Arévalo M.
;
Ayala-Chauvin, Manuel Ignacio
;
Nacevilla C.
;
de la Fuente-Morato A.
At present, the handcrafted footwear sector is affected by the high competitiveness due to the increasing automation of companies. In this sense, in order to improve its competitiveness, a system was proposed to predict the failures of a production system and to carry out preventive maintenance actions. Samples were taken from 25 productions and 7 activities were established: cutting, stitching, pre fabrication, final preparation, gluing, assembly and finishing. The company produces batches of 90 pairs per day, with a standard time of 274.53 min and a promised productivity of 1.8. A support vector machine model was developed to predict the possible failures of the process taking as a reference the standard time of each stage. Finally, the results allow predicting the faults to optimise the production process by applying Support Vector Machine (SVM). © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
25

Ayala-Chauvin, Manuel Ignacio

Filters

Settings

Sort By

Results per page