Ciencias Exactas y Ciencias de la Salud
Permanent URI for this collectionhttps://hdl.handle.net/11285/551014
Pertenecen a esta colección Tesis y Trabajos de grado de los Doctorados correspondientes a las Escuelas de Ingeniería y Ciencias así como a Medicina y Ciencias de la Salud.
Browse
Search Results
- A novel feature extraction methodology using Inter-Trial Coherence framework for signal analysis – A case study applied towards BCI(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2024-11) López Bernal, Diego; Ponce Cruz, Pedro; emipsanchez; Ponce Espinosa, Hiram; López Caudana, Edgar Omar; Bustamante Bello, Martín Rogelio; School of Engineering and Sciences; Campus Ciudad de México; Balderas Silva, David ChristopherSignal classification in environments with low signal-to-noise ratio (SNR) presents a significant challenge across various fields, from industrial monitoring to biomedical appli cations. This work explores a novel methodology aimed at improving classification accuracy in such conditions, using EEG-based Brain-Computer Interfaces (BCIs) for inner speech decoding as a case study. EEG-based Brain-Computer Interfaces (BCIs) have emerged as a promising technology for providing communication channels for individuals with speech disabilities, such as those affected by amyotrophic lateral sclerosis (ALS), stroke, or other neurodegenerative diseases. Inner speech classification, a subset of BCI applications, aims to interpret and translate silent, inner speech into meaningful linguistic information. De spite the potential of BCIs, current methodologies for inner speech classification lack the accuracy needed for practical applications. This work investigates the use of inter-trial coherence (ITC) as a novel feature extraction technique to enhance the accuracy of in ner speech classification in EEG-based BCIs. The study introduces a methodology that integrates ITC within a complex Morlet time-frequency representation framework. EEG recordings from ten participants imagining four distinct words (up, down, right, and left) were processed and analyzed. Five different classification algorithms were evaluated: Ran dom Forest (RF), Support Vector Machine (SVM), k-Nearest Neighbors (kNN), Linear Discriminant Analysis (LDA), and Naive Bayes (NB). The proposed method achieved no table classification accuracies of 75.70% with RF and 66.25% with SVM, demonstrating significant improvements over traditional feature extraction methods. These findings indi cate that ITC is a viable technique for enhancing the accuracy of inner speech classification in EEG-based BCIs. The results suggest practical implications for improving communica tion and navigation capabilities for individuals with ALS or similar conditions. This work lays the foundation for future research on phase-based feature extraction, opening new avenues for understanding the neural mechanisms underlying inner speech and advancing BCI systems’ accuracy and efficiency
- 3D Computer vision for online activity detection. Case study: metabolic rate estimation for connected thermostat(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2022-12-01) Mata Juárez, Omar; PONCE CRUZ, PEDRO; 31857; Ponce Cruz, Pedro; puemcuervo, emipsanchez; Peffer, Therese; McDaniel, Troy; López Caudana, Edgar Omar; School of Engineering and Sciences; Campus Monterrey; Molina Gutiérrez, ArturoThe ability to detect human activities in computer vision has gained importance over the years due to its potential in many applications such crime prevention, healthcare, public safety, human-computer/robot interaction, smart homes, videogames, monitoring, etc. A way to achieve those applications is by doing a Human Activity Recognition (HAR) process in which an activity is identified by a series of physical actions that construct one physical activity. The identification requires sensors to obtain the data for processing and classifying it. These kinds of sensors are often found inside a smart home. Therefore, it is proposed to use noninvasive sensors in combination with digital signal processing to develop a platform for detecting human activity. Moreover, a case study is proposed for validating the platform by proposing a strategy to save energy on HVAC systems without affecting the thermal comfort of the occupant
- Tailored gamification platform based on artificial intelligence. Connected thermostats as a case study for saving energy in connected homes(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2022-12) Méndez Garduño, Juana Isabel; MENDEZ GARDUÑO, JUANA ISABEL; 686197; Ponce Cruz, Pedro; emipsanchez; McDaniel, Troy; López Caudana, Edgar Omar; Molina Gutiérrez, Arturo; School of Engineering and Sciences; Campus Monterrey; Peffer, ThereseThe product platforms are a set of system components that are interdependent with other system components. Furthermore, platforms are the basis for all technology-based products and allow collaborations for multi-product systems. Traditionally, products were created without third-party collaboration. Thus, the same owner's product upgraded, modified, or updated the product falling in limited customization, lack of integration, and modularity. Evolving products into product platforms creates value, but it is complex to implement. The relevance of transitioning into product platforms relies on companies entering global markets. Therefore, platforms are cost-effective for global competition. For instance, around 60~\% of technological companies value investing in platforms. Furthermore, the tendency shows that companies aspire to turn the business into a fully integrated digital technology company. On the other hand, customers prefer a tailored service, platform, or product over generic products. Nevertheless, the adoption of these product platforms fails due to usability and behavioral problems. Hence, it is complex to measure individuals’ satisfaction because their behavior is related to perception and other context-specific factors, such as age, gender, income, cultural aspects, specific needs, personality traits, and other preferences. To achieve the adoption of product platforms, this thesis proposes to tailor user solutions by profiling the consumer through personality traits to propose strategies that allow them to adapt more easily to product usage. Thus, appealing ludic interfaces engage end-users to interact better with platforms. Therefore, social interaction (social platform) plays a primary role in understanding and knowing better the users’ patterns and profiles them. In addition, it is feasible to understand consumers' habits by sending stimuli through gamification or serious game strategies. Gamification enhances a platform with affordances for gameful experiences to support the user’s overall value creation. Besides, Artificial Intelligence decision systems link the type of consumer and gamification for deploying user-oriented product platforms. Hence, this thesis proposed a four-step methodology for deploying tailored platforms and validating the methodology in a case study. Furthermore, this methodology was used in the context of smart homes, smart communities, and smart cities.
- Analysis and use of textual definitions through a transformer neural network model and natural language processing(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2021-12-02) Baltazar Reyes, Germán Eduardo; BALTAZAR REYES, GERMAN EDUARDO; 852898; Ponce Cruz, Pedro; puemcuervo; McDaniel, Troy; Balderas Silva, David Christopher; Rojas Hernández, Mario; School of Engineering and Sciences; Campus Ciudad de México; López Caudana, Edgar OmarThere is currently an information overload problem, where data is excessive, disorganized, and presented statically. These three problems are deeply related to the vocabulary used in each document since the usefulness of a document is directly related to the number of understood vocabulary. At the same time, there are multiple Machine Learning algorithms and applications that analyze the structure of written information. However, most implementations are focused on the bigger picture of text analysis, which is to understand the structure and use of complete sentences and how to create new documents as long as the originals. This problem directly affects the static presentation of data. For these past reasons, this proposal intends to evaluate the semantical similitude between a complete phrase or sentence and a single keyword, following the structure of a regular dictionary, where a descriptive sentence explains and shares the exact meaning of a single word. This model uses a GPT-2 Transformer neural network to interpret a descriptive input phrase and generate a new phrase that intends to speak about the same abstract concept, similar to a particular keyword. The validation of the generated text is in charge of a Universal Sentence Encoder network, which was finetuned for properly relating the semantical similitude between the total sum of words of a sentence and its corresponding keyword. The results demonstrated that the proposal could generate new phrases that resemble the general context of the descriptive input sentence and the ground truth keyword. At the same time, the validation of the generated text was able to assign a higher similarity score between these phrase-word pairs. Nevertheless, this process also showed that it is still needed deeper analysis to ponderate and separate the context of different pairs of textual inputs. In general, this proposal marks a new area of study for analyzing the abstract relationship of meaning between sentences and particular words and how a series of ordered vocables can be detected as similar to a single term, marking a different direction of text analysis than the one currently proposed and researched in most of the Natural Language Processing community.
- Passive decentralized island mode detection and optimization-based design of passive filters for disconnection events in microgrid systems(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2021-12-06) López Gutiérrez, Juan Roberto; Ponce Cruz, Pedro; puemcuervo; Balderas Silva, David Christopher; Reyes Rosario, Alfredo; Soriano Avedaño, Luis Arturo; School of Engineering and Sciences; Campus Ciudad de México; Ibarra Moyers, Luis MiguelIn recent years, the electrical network has been evolving towards becoming a more sustainable system, the present environmental concerns regarding the greenhouse gas emission by the energy sector have pushed forward the integration of alternative generation units that promote the decarbonization of the energy production sector. Over the past decades the integration of these ``cleaner'' energy generation systems has been done in an On-grid and an Off-grid fashion, however, this integration strategy encountered some problems regarding key areas such as control and management, just to mention a few. The microgrid concept is then created to overcome these issues, allowing a seamless integration to the electrical network of the growing alternative generation assets, improving how these ``cleaner'' energy production alternatives are managed into more sustainable systems. In Microgrids with a high penetration of renewable energy sources, power converters are used to regulate the produced energy to a single voltage and frequency reference value across the microgrid. Adequate incorporation of an LC filter at the output of power electronic devices allows the attenuation of harmful harmonics that can be introduced to the microgrid's energy bus. By traditional methods, LC filter values can be calculated by means of the power rating, switching frequency, cutoff frequency, and using the bode frequency domain. It is important to consider that, a microgrid including distributed generators can operate connected to the main electrical network or in an isolated manner, referred to as island operation. The transition between both states can occur voluntarily, but a disconnection can also happen unexpectedly. The associated transients can be harmful to the grid, and compensating actions must be triggered to avoid service interruption, preserve power quality, and minimize the possibility of faults. It is important to consider that in transition from a connected to an autonomous microgrid operation, the calculated LC filter can lead to high harmonic injection. As a result, a tuning methodology capable of obtaining the right set of parameters for the LC filter for such transition events can improve the performance of the microgrid. Alternately, such transition events must be detected to enable compensating action; island detection methods are essential to this end. Such techniques typically depend on communication networks or on the introduction of minor electrical disturbances to identify and broadcast unexpected islanding events. However, local energy resources are distributed, variable, and are expected to be integrated in a plug-and-play manner; then, conventional island detection strategies can be ineffective as they rely on specific infrastructure. To overcome those problems, this work proposes to improve the islanding phenomenon in two main contributions. To tackle the issues in regards to the introduction of harmful transients by traditional LC filters, this work optimizes the LC output parameters with respect to the size of the filter components, the IEEE Std 519-2014, and bandwidth of the filter, within a bounded region of values subjected to performance conditions such as voltage output, and the produced total harmonic distortion measurements during the transition from a connected to an autonomous operation. In a case study, genetic algorithm optimization is used to obtain the LC filter parameters and compared to a conventional arithmetic methodology to obtain the values of the filter. The optimization results in a set of values that lead to a higher harmonic attenuation after the transition rather than a conventional method using the switching frequency as the main design factor. In the other end of the islanding phenomenon, where islanding events must be detected while avoiding traditional infrastructure setbacks, a straightforward, distributed island detection technique is proposed, this technique relies only on local electrical measurements, available at the output of each generating unit. The proposed method is based on the estimated power-frequency ratio, associated with the stiffness of the grid. A ``stiffness change'' effectively reveals island operating conditions, discards heavy load variations, and enables independent (distributed) operation. The proposal was validated through digital simulations and an experimental test-bed. Such test-bed consists of a Real-Time HIL implementation, the proposed island detection algorithm is programmed to run in an embedded format while connected to a Real-Time simulator running a microgrid equivalent model in the form of a three-phase parallel RLC load as recommended by the IEEE Std. 929 and IEEE Std. 1547 for islanding detection. Results showed that the proposed technique can effectively detect island operation at each generating unit interacting in the microgrid. Moreover, it was about three times faster than other reported techniques.
