Luis Felipe Marín Urías

Non-invasive blood glucose level (BGL) estimation using photoplethysmography has been studied as a promising alternative to enzyme-based technologies. However, the reliability of current machine learning models is frequently compromised by overfitting and data leakage. This paper presents a methodological validation framework designed to expose and mitigate these issues by objectively comparing spectral and temporal statistical feature extraction strategies using the MUST dataset. The signals were pre-processed and evaluated through two distinct approaches: extracting 12 Mel-frequency cepstral coefficients (MFCC) for a support vector regressor (SVR), and 22 time-domain statistical features for random forest (RF) and Extreme Gradient Boosting (XGBoost) models. To expose data leakage, models were rigorously tested using both standard 5-fold cross-validation and a strict subject-independent grouped validation. Experimental results confirm that standard cross-validation inflates predictive metrics; the MFCC-SVR model initially achieved a moderate correlation ( r = 0.4818 , RMSE = 17.00 mg dl − 1 ) under this random-split scheme. However, when evaluated through the proposed subject-independent validation framework, the predictive correlation dropped significantly ( r = 0.3618 , RMSE = 17.44 mg dl − 1 ). This degradation demonstrates that the models explain a low percentage of actual BGL variance and tend to output predictions near the population mean rather than tracking individual physiological fluctuations. While the MFCC approach proved comparatively more stable than temporal statistical models (RF: r = 0.1904 ; XGBoost: r = 0.1645 ), its weak predictive correlation indicates it is not currently suitable as a standalone tool for clinical decision support. Instead, while 100% of predictions fall within Zones A and B of the Clarke Error Grid Analysis, this outcome reflects the limited blood glucose distribution of the dataset rather than clinical deployment readiness. Ultimately, by exposing the limitations of current models under strict validation, this study establishes the critical need for larger, comprehensive datasets to enable the development of truly robust non-invasive monitoring technologies.

A control approach for linear actuators with time delay is introduced. The novelty of the proposal is the inclusion of a virtual reference trajectory (VRT) without modifying the controller architecture, allowing the stability margin of a single-delay actuator controller to be increased; in the ideal case, the proposed approach can render stability independent of delay. Moreover, a slight modification of the dynamics of the virtual trajectory allows the closed-loop system to be expressed as a neutral time-delay system. Therefore, its stability analysis is performed using the Lyapunov–Krasovskii approach, yielding theoretical conditions that allow the critical time delay to be computed. Numerical simulations and experiments on the control of a DC motor actuator with a control input time delay validate our proposal.

Este libro presenta una revisión exhaustiva del análisis y aplicación práctica del analizador vectorial de redes NanoVNA S-A-A-2 en el estudio de las líneas de transmisión y circuitos de RF. Destaca su utilidad como herramienta didáctica y de investigación en el aprendizaje y desarrollo de habilidades prácticas en dicha área. Los colaboradores son profesores de la Facultad de Ingeniería Eléctrica y Electrónica y, en su mayoría, integrantes de dos cuerpos académicos actualmente consolidados. La obra aporta una metodología estructurada de validación experimental, que puede ser adoptada para investigaciones y prácticas de enseñanza, para enriquecer la formación en el área de RF a estudiantes y profesionales. Se detalla el paso a paso para la calibración, configuración de montajes y análisis de resultados usando el NanoVNA, lo cual le permite al lector reproducir y adaptar las técnicas en diferentes entornos de laboratorio. Además, se fomenta el aprendizaje activo mediante la resolución de ejemplos prácticos y ejercicios orientados a aplicaciones profesionales y académicas. Entre las principales aportaciones de esta obra frente a otras bibliografías, destaca la integración del equipo NanoVNA S-A-A-2 en un esquema pedagógico asequible y accesible, que facilita el aprendizaje y la experimentación con instrumentos de bajo costo sin sacrificar precisión, fomentando el desarrollo de razonamientos críticos en la resolución de problemas en el área de RF.

An approach for the control of linear control systems with a single time-delay is proposed. The main contribution is the inclusion of a symmetric-injection virtual reference trajectory into the controller to render stability robustness of single-delay linear control systems. The dynamics of the virtual trajectory is included into the closed-loop dynamics allowing theoretical computation of the critical time-delay before losing stability. Moreover, an energy-based symmetry interpretation of the proposed approach is drawn. Numerical simulations considering stable and unstable linear systems are shown, and experiments to control a DC-motor with time-delay measurements validate our proposal.

Biogas production estimation has been one of the most important and challenging objectives for anaerobic digestion processes due to the complexity of its dynamics and the lack of high-quality open-access datasets. This study presents a hybrid modeling framework that combines a mechanistic model, based on ordinary differential equations (ODE), with a machine learning model. Rather than relying exclusively on experimental data, the proposed approach leverages physics-informed synthetic data generation, complemented by a lag-based feature engineering to capture inherent temporal dependencies in the process dynamics available in operational data of a bio-digester. Two configurations were evaluated: a baseline model and an enhanced version incorporating lag features and simplified temperature profile. While the improved model achieved high predictive performance (R2=0.97885, RMSE=131.80[L/d]), additional analyses reveal that this performance is partly driven by temporal memory and remains sensitive to noise and feature composition. Instead of presenting the model as a final solution, this work frames it as a step toward practical digital twin implementations, acknowledging the gap that still exists between simulation-based accuracy and real-world reliability.

Enhancement and restoration algorithms are widely used in the exploration of coral reefs for improving underwater images. However, by selecting an improvement algorithm based on image quality metrics, image processing key factors such as the execution time are not considered. In response to this issue, herein is presented a novel method built on multicriteria decision analysis that evaluates the processing time and feature point increase with respect to the original image. To set the Decision Matrix (DM), both the processing time and keypoint increase criteria of the evaluated algorithms are normalized. The criteria weights in the DM are set in accordance with the application, and the quantitative metric used to select the best alternative is the highest Weighted Sum Method (WsuM) score. In this work, the DM of six scenarios is shown, since the setting of weights could completely change the decision. For this research’s target application of generating underwater photomosaics, the Dark Channel Prior (DCP) algorithm emerged as the most suitable under a weighting scheme of 75% for processing time and 25% for keypoint increase. This proposal represents a solution for evaluating improvement algorithms in applications where computational efficiency is critical.

For years, efforts have been devoted to establishing an effective bull breeding soundness evaluation procedure; usual research on this subject is based on bull breeding soundness examination (BBSE) methodologies, which have significant limitations in terms of their evaluation procedure, such as their high cost, time consumption, and administrative difficulty, as well as a lack of diagnostic laboratories equipped to handle the more difficult cases. This research focused on the creation of a prediction model to supplement and/or improve the BBSE approach through the study of two algorithms, namely, clustering and artificial neural networks (ANNs), to find the optimum machine learning (ML) approach for our application, with an emphasis on data categorization accuracy. This tool was designed to assist veterinary medicine and farmers in identifying key factors and increasing certainty in their decision-making during the selection of bulls for breeding purposes, providing data from a limited number of factors generated from a deep pairing study of bulls. Zebu, European, and crossbred bulls were the general groupings. The data utilized in the model’s creation (N = 359) considered five variables that influence improvement decisions. This approach enhanced decision-making by 12% compared to traditional breeding bull management. ANN obtained an accuracy of 90%, with precision rates of 97% for satisfactory, 92% for unsatisfactory, and 85% for bad. These results indicate that the proposed method can be considered an effective alternative for innovative decision-making in traditional BBSE.

For years, effort has been devoted to establishing an effective bull breeding soundness evaluation procedure; usual research on the subject is based on BBSE (Bull Breeding Soundness Examination) methodologies, which have significant limitations in terms of their evaluation procedure, such as high cost, time-consuming and administratively difficult. This research focuses on the creation of a prediction model to supplement and/or improve the BBSE approach, through the study of two algorithms, Clustering and Artificial Neural Network (ANN), to find the optimum Machine Learning (ML) approach for our application, with an emphasis on data categorization accuracy. The tool is designed to assist veterinary medicine and farmers in identifying key factors and increasing certainty in their decision-making during the selection of bulls for breeding purposes, providing data from a limited number of factors generated from a deep pairing study of bulls. Zebu Bulls, European Bulls, and Crossbred Bulls were the general groupings. The data utilized in the model's creation (N =359) considers five variables that influence improvement decisions. This approach enhances decision-making by 12%. ANN obtained an accuracy of 90%; Precision was 97% for satisfying, 92% for unsatisfying, and 85% for Bad.