Ciencias Exactas y Ciencias de la Salud
Permanent URI for this collectionhttps://hdl.handle.net/11285/551039
Pertenecen a esta colección Tesis y Trabajos de grado de las Maestrías correspondientes a las Escuelas de Ingeniería y Ciencias así como a Medicina y Ciencias de la Salud.
Browse
Search Results
- Improving deep neural networks to identify depression using neural architecture search(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2025-06) Hernández Silva, Erick; Trejo Rodríguez, Luis Ángel; emipsanchez; Cantoral Ceballos, José Antonio; González Mendoza, Miguel; School of Engineering and Sciences; Campus Estado de México; Sosa Hernández, Víctor AdriánA Neural Architecture Search (NAS) framework utilizing Evolutionary Algorithms (EAs) and a regressor model is proposed to improve the classification performance of Deep Neural Net- works (DNNs) for the early detection of Major Depressive Disorder (MDD) from speech data represented by Mel-Spectrograms. The framework automates the design of neural network architectures by systematically exploring a well-defined search space that integrates convo- lutional layers, batch normalization, dropout, max pooling, and self-attention mechanisms, aiming to capture both spatial and temporal features inherent in vocal signals. By optimiz- ing for the F1-score, the framework addresses challenges related to data imbalance, ensuring robust generalization across both depressed and non-depressed samples. The proposed approach employs an integer-based encoding scheme to represent candi- date architectures, coupled with repair and validation processes that ensure all architectures meet specific design constraints. A self-adaptive mechanism dynamically adjusts the muta- tion factor based on evolutionary feedback, improving the balance between exploration and exploitation during the search process. Furthermore, a surrogate model, built using Princi- pal Component Analysis (PCA) and XGBoost regressor, predicts architecture performance, significantly reducing computational costs by avoiding full model training for all candidates. Experimental validation, conducted on publicly available speech datasets, demonstrates that NAS-generated architectures may outperform manually designed state-of-the-art models in terms of F1-score, accuracy, precision, recall, and specificity. The results confirm the effec- tiveness of integrating self-attention mechanisms with convolutional operations for extracting relevant depression-related vocal biomarkers. This research underlines the potential of NAS in advancing non-invasive, scalable, and interpretable AI-driven tools for mental health as- sessment, contributing to early intervention strategies and improving clinical outcomes in depression diagnosis.
- D3TEC Dataset: a data collection for deep learning research in depression classification featuring voice recordings of Spanish speakers using professional and cellphone microphones(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2024-05) Brenes García, Luis Felipe; Trejo Rodríguez, Luis Ángel; emimmayorquin; Villaseñor Pineda, Luis; Sosa Hernández, Víctor Adrián; School of Engineering and Sciences; Campus Monterrey; Cantoral Ceballos, José AntonioDepression is a mental health condition that affects millions of people worldwide. Although common, it remains difficult to diagnose due to its heterogeneous symptomatology. Mental health questionnaires are currently the most used assessment method to screen depression; these, however, have a subjective nature due to their dependence on patients' self-assessments. Researchers have been interested in finding an accurate way of identifying depression through an objective biomarker. Recent developments in neural networks and deep learning have enabled the possibility of classifying depression through the computational analysis of voice recordings. However, this approach is heavily dependent on the availability of datasets to train and test deep learning models, and these are scarce. There are also very few languages available. This study proposes a protocol for the collection of a new dataset for deep learning research on voice depression classification, featuring Spanish speakers, professional and smartphone microphones, and a high-quality recording standard. This work aims at creating a high-quality voice depression dataset by recording Spanish speakers with a professional microphone and strict audio quality standards. The data is captured by a smartphone microphone as well for further research in the use of smartphone applications for depression identification. Our methodology involves the strategic collection of depressed and non-depressed voice recordings. Three types of data are collected: voice recordings, depression labels (using the PHQ-9 questionnaire), and additional data that could potentially influence speech. Recordings are captured with professional-grade and smartphone microphones simultaneously to ensure versatility and practical applicability. Several considerations and guidelines are described to ensure high audio quality and avoid potential bias in deep learning research. This data collection effort immediately enables new research topics on depression classification. Some potential uses include deep learning research on Spanish speakers, an evaluation of the impact of audio quality on developing audio classification models, and an evaluation of the applicability of voice depression classification technology on smartphone applications. A preliminary experimentation section is included to showcase the potential research areas that the creation of this dataset enables. This research marks a significant step towards the objective and automated classification of depression in voice recordings. By focusing on the underrepresented demographic of Spanish speakers, the inclusion of smartphone recordings, and addressing the current data limitations in audio quality, this study lays the groundwork for future advancements in deep learning-driven mental health diagnosis.
- BMV stocks return prediction using macro economics variables, technical analysis, and machine learning(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2020-04-01) Hinojosa Alejandro, Ramón; TREJO RODRIGUEZ, LUIS ANGEL; 59028; Trejo Rodríguez, Luis Ángel; puemcuervo; Hervet Escobar, Laura; School of Engineering and Sciences; Campus Monterrey; Hernández Gress, NeilHistorical data, macroeconomic variables, technical analysis, and machine learning are some of the tools used to predict the price of shares of companies listed on the Mexican stock ex-change.The present thesis’s purpose is to reach a robust investment strategy, capable of coping with unforeseen events, and maximizing returns by selecting stocks quoted in the Mexican Stock Market. Our strategy predicts stock returns considering the influence of macroeconomic variables filtered by a causal analysis to determine the most significant ones, and a layered architecture, where machine learning methodologies are endowed with technical analysis applied to the stock historical data.The results from this thesis work show profitable strategies that outperform the free-risk rate of return and the Mexican Index performance. Results demonstrate even good performances when unforeseen events are present as the Covid-19 pandemic in 2020-2021.
- Estudio para la implantación de una autoridad certificadora y su entorno(Instituto Tecnológico y de Estudios Superiores de Monterrey, 1999) Perea Páez, Sergio Arturo; SERGIO ARTURO PEREA PÁEZ; Trejo Rodríguez, Luis Ángel; Gómez C., Roberto; Vázquez, JesusEste trabajo de tesis tiene como finalidad servir de guía para la implantación de una infraestructura de certificación para las organizaciones mexicanas, o en su caso poder montar una Autoridad Certificadora independiente. En este trabajo se establecen los elementos necesarios para poder montar, administrar y operar una Autoridad Certificadora. Así mismo se especifican los Procesos de Certificación que debe seguir una Autoridad Certificadora para poder funcionar de manera adecuada. Primeramente, se hace la presentación de la teoría de Certificados Digitales. Esta teoría consiste en los fundamentos criptográficos de los Certificados Digitales, los cuales están dados en la criptografía de llave pública. Posteriormente se hace un estudio de los Certificados Digitales, desde el punto de vista criptográfico. Como una de las partes más importantes de este trabajo, se plantea el cómo montar, administrar y operar una Autoridad Certificadora y se establecen sus Procesos de Certificación. Una vez expuesta la teoría acerca de la Autoridad Certificadora, ésta se adecua a la operación cotidiana de una organización y se utiliza una herramienta comercial (Servidor de Certificados de Netscape) como una opción de software práctica para implementar una infraestructura de Certificación. Con la finalidad de resolver el problema de certificación del ITESM-CEM y con el objeto de servir como una aplicación práctica de este trabajo de tesis, se presenta el "Caso de estudio del ITESM-CEM", en el cual se busca implementar una infraestructura de Certificación a gran escala. También se hace una breve presentación del protocolo SSL por su estrecha relación con el uso de Certificados Digitales. Y por último se presenta la reseña del trabajo práctico que sustenta gran parte de este trabajo de tesis, el cual fue realizado en el Departamento de Ciencias Computacionales del ITESM-CEM.