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
- Neutrino classification through deep learning amid the Hyper-Kamiokande project development(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2024-06-10) Romo Fuentes, María Fernanda; Falcón Morales, Luis Eduardo; emipsanchez; Cuen Rochin, Saul; De la Fuente Acosta, Eduardo; School of Engineering and Sciences; Campus Estado de MéxicoNeutrinos are a type of elemental particle that are characterized by the fact that their mass is really small, that they have no electric charge and present a special behavior called oscillation in which they can be measured to be of a kind different to the one they actually are. All these characteristics make neutrinos one of the most studied particles by different researchers and in different facilities nowadays, since the information we can obtain from its study allows us to solve some of the Universe’s greatest mysteries. One of these projects where neutrinos are studied is the Hyper-Kamiokande which refers to both, the international collaboration of researchers, to which Mexico belongs to, and the neutrino grand-scale detector based on Cherenkov radiation currently being built in Japan. In this detector the data of a neutrino event is collected by a special kind of sensors located in its walls called Photo Multiplier Tubes or PMTs, to then be analyzed, and this analysis usually starts by the identification of the particles involved in an event, which is where this project comes forth, since an appropriate method to classify neutrinos based on the radiation pattern they leave as they pass through the detector is needed. Hence, in the following project to obtain the Master in Computer Science degree, we implement and test 4 deep learning architectures: VGG19, ResNet50, PointNet and Vision Transformer, for the classification of neutrinos since they are state of the art methods, this is, they are architectures used as the starting point for any classification task and, moreover, we can tune them and/or apply different techniques such as regularization to get the best possible performance while reducing overfitting. Using the mentioned architectures we process a dataset composed of neutrino events simulated by a software called WCSim in 2021. These events are of single ring type, correspond to the IWCD tank, a smaller tank being built to aid in the tasks of the Hyper-Kamiokande and range from 9 thousand to 8 million per each of the 3 particles considered in the project: muon and electron neutrinos and gamma particles. The results show that ResNet50 was the architecture that gave the best results while also minimizing the computational resources needed, though its performance is similar to the one given by VGG19 and PointNet, they require a larger time to process any dataset, whereas Vision Transformer provided the poorest results, however, all results improved by processing the largest datasets. Then, in comparison with a state of the art custom CNN we found that our highest average accuracy is within the same range as the one they obtained, whereas, in comparison with the ResNet50 model currently being used in the HK collaboration we found that the obtained AUC for the TPR signal (electron) vs FPR background (gamma) curve for our best model is 0.71, whereas this AUC value for the collaboration is 0.77, nonetheless, we have to consider that to obtain this value the whole results are not analyzed by the collaboration but cuts are applied and therefore, our results can be considered close.
- Deep Learning Approach for Alzheimer’s Disease Classification: Integrating Multimodal MRI and FDG- PET Imaging Through Dual Feature Extractors and Shared Neural Network Processing(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2024) Vega Guzmán, Sergio Eduardo; Alfaro Ponce, Mariel; emimmayorquin; Ochoa Ruíz, Gilberto; Chairez Oria, Jorge Isaac; Hernandez Sanchez, Alejandra; School of Engineering and Sciences; Campus Monterrey; Ramírez Nava, Gerardo JuliánAlzheimer’s disease (AD) is a progressive neurodegenerative disorder whose incidence is expected to grow in the coming years. Traditional diagnostic methods, such as MRI and FDG-PET, each provide valuable but limited insights into the disease’s pathology. This thesis researches the potential of a multimodal deep learning classifier to improve the diagnostic accuracy of AD by integrating MRI and FDG-PET imaging data in comparison to single modality implementations. The study proposes a lightweight neural architecture that uses the strengths of both imaging modalities, aiming to reduce computational costs while maintaining state-of-the-art diagnostic performance. The proposed model utilizes two pre-trained feature extractors, one for each imaging modality, fine-tuned to capture the relevant features from the dataset. The outputs of these extractors are fused into a single vector to form an enriched feature map that better describes the brain. Experimental results demonstrate that the multimodal classifier outperforms single modality classifiers, achieving an overall accuracy of 90% on the test dataset. The VGG19 model was the best feature extractor for both MRI and PET data since it showed superior performance when compared to the other experimental models, with an accuracy of 71.9% for MRI and 80.3% for PET images. The multimodal implementation also exhibited higher precision, recall, and F1 scores than the single-modality implementations. For instance, it achieved a precision of 0.90, recall of 0.94, and F1-score of 0.92 for the AD class and a precision of 0.89, recall of 0.82, and F1-score of 0.86 for the CN class. Furthermore, explainable AI techniques provided insights into the model’s decisionmaking process, revealing that it effectively utilizes both structural and metabolic information to distinguish between AD and cognitively normal (CN) subjects. This research adds supporting evidence into the potential of multimodal imaging and machine learning to enhance early detection and diagnosis of Alzheimer’s disease, offering a cost-effective solution suitable for widespread clinical applications.
- Road surface monitoring system through machine learning classification ensemble models(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2022-12) Arce Sáenz, Luis Alejandro; Bustamante Bello, Martin Rogelio; puelquio, emipsanchez; Villagra Serrano, Jorge; Galluzzi Aguilera, Renato; Ramírez Mendoza, Ricardo Ambrocio; School of Engineering and Sciences; Campus Ciudad de México; Izquierdo Reyes, JavierThe development of megacities is currently the scene of many problems; an important one to consider is the quality and efficiency of their mobility. An essential factor impacting this is the quality of their road networks, which can affect the durability and safety of ground transportation systems. Mexico City is a great example of such deficiencies. Therefore smart mobility strategies and planning in terms of logistics have been proposed, but few technological integrations have been implemented. In this work, a platform capable of monitoring surface defects in road pavement using Inertial Measurement Units and Machine Learning classification models was designed and developed. This was achieved by recording accelerometer and gyroscope measurements on a test vehicle's damped and undamped mass while driving on Mexico City streets. The measurements were labeled to identify and classify general and specific elements of road irregularities: smooth and uneven road segments, potholes, manholes, speed bumps, and patches. It is described as a methodology for preprocessing the data through time series analysis and feature extraction in the time and frequency domain. Four ensemble models were trained using the best classification models out of eight candidates; an exhaustive grid search methodology was used to select the best classification models per category and optimize the system's performance. Finally, the algorithms and models were loaded into a cloud instance to process incoming raw data; the resultant predictions were stored in a cloud database to be visualized on a web platform.
- Deep learning for clothing classification, case study:thermal comfort(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2021-11-23) Medina Rosales, Adán; Ponce Cruz, Pedro; puemcuervo; López Caudana, Edgar Omar; Rojas Hernández, Mario; Soriano Avendaño, Luis Arturo; School of Engineering and Sciences; Campus Ciudad de México; Molina Gutiérrez, ArturoImage classification algorithm has being in quick development over the last 10 years with a new algorithm appearing every year, this new algorithms aim to be faster and more accurate than its predecessors, so real time implementations for object classifiers are more frequent. However the solutions for problems are going to more complex problems leaving things such as clothing ensemble classification on the side. There are some proposed solutions on the recognition of clothing garments but all aim to a specific solution in the fashion industry for customer categorization or shopping proposals, however a more general approach which recognizes multiple clothing garments is missing, and a real time clothing ensemble detection could be implemented in several problems. One of such problems is the case study for this project were a CNN implementation is used in video testing to propose the solution for clothing insulation determination using the real time clothing ensemble detector and therefore have a more accurate thermal comfort value. The results proved that the implementation of the chosen CNN architecture could be used as a clothing ensemble detector in a real time implementation, however since a minimized version of the needed dataset was used to verify the viability of this proposal a more complete dataset needs to be created in order to improve the models performance. In general this proposal shows the comparison between come CNN architectures and the datasets available for the propose objectives, as well as the creation of a new dataset that can be successfully used to train the chosen CNN model and produce a real time clothing ensemble detector.
- Detection of suspicious attitudes on video using neuroevolved shallow and deep neural networks models(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2021-11) Flores Munguía, Carlos; Terashima Marín, Hugo; puemcuervo/tolmquevedo; Oliva, Diego; Ortiz Bayliss, Jose Carlos; School of Engineering and Sciences; Campus MonterreyThe analysis of surveillance cameras is a critical task usually limited by the people involved in the video supervision devoted to such a task, their knowledge, and their judgment. Security guards protect other people from different events that can compromise their security, like robbery, extortion, fraud, vehicle theft, and more, converting them to an essential part of this type of protection system. If they are not paying attention, crimes may be overlooked. Nonetheless, different approaches have arisen to automate this task. The methods are mainly based on machine learning and benefit from developing neural networks that extract underlying information from input videos. However, despite how competent those networks have proved to be, developers must face the challenging task of defining the architecture and hyperparameters that allow the network to work adequately and optimize the use of computational resources. Furthermore, selecting the architecture and hyperparameters may significantly impact the neural networks’ performance if it is not carried out adequately. No matter the type of neural network used, shallow, dense, convolutional, 3D convolutional, or recurrent; hyperparameter selection must be performed using empirical knowledge thanks to the expertise of the designer, or even with the help of automated approaches like Random Search or Bayesian Optimization. However, such methods suffer from problems like not covering the solution space well, especially if the space is made up of large dimensions. Alternatively, the requirement to evaluate the models many times to get more information about the evaluation of the objective function, employing a diverse set of hyperparameters. This work proposes a model that generates, through a genetic algorithm, neural networks for behavior classification within videos. The application of genetic algorithms allows the exploration in the hyperparameters solution space in different directions simultaneously. Two types of neural networks are evolved as part of the thesis work: shallow and deep networks, the latter based on dense layers and 3D convolutions. Each sort of network takes distinct input data types: the evolution of people’s pose and videos’ sequences, respectively. Shallow neural networks are generated by NeuroEvolution of Augmented Topologies (NEAT), while CoDeepNEAT generates deep networks. NEAT uses a direct encoding, meaning that each node and connection in the network is directly represented in the chromosome. In contrast, CoDeepNEAT uses indirect encoding, making use of cooperative coevolution of blueprints and modules. This work trains networks and tests them using the Kranok-NV dataset, which exhibited better results than their competitors on various standard metrics.
- Development and implementation of a categorization model for the exoskeletons based on their design characteristics and practical projects(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2020) de la Tejera de la Peña, Javier Alberto; BUSTAMANTE BELLO, MARTIN ROGELIO; 58810; IZQUIERDO REYES, JAVIER; 710170; Bustamante Bello, Martín Rogelio; emipsanchez; Izquierdo Reyes, Javier; School of Engineering and Sciences; Campus Ciudad de México; Ramírez Mendoza,Ricardo AmbrocioThe exoskeletons are the future of the humankind. The humankind is in a constant pursuit of improvement for themselves, both physically and mentally. The human body has physical constraints in which no physical training can surpass, but our ingenious and imagination are making this possible. Several decades ago the first developments started, and nowadays, these developments, plus the improvements, are imperative for the humans in the following decades. The exoskeletons can assist or rehabilitate a person, leading this personalized technology to depend on the needs and abilities that each user has. The exoskeletons have a wide spectrum of opportunities in their design, due to the variety of situations in which a person needs an augmentation of their physical performance, thus the diversity of projects. An exoskeleton for sarcopenia was made for assisting the elderly who require help to perform their daily activities, and tested with electromyography (EMG) to analyze its functionality. On the other hand, an exoskeleton made for rehabilitation, machining the exoskeleton gives us a testing platform for other kinds of projects. Through the development of different exoskeletons and projects related to them, an opportunity area was found to formalize the exoskeletons’ topic, creating a model for the categorization of all the exoskeletons using their design characteristics and a further analysis for recommendations in their design. Besides, in this work are proposed tools, based on the design characteristics of exoskeletons, for the optimization of the exoskeleton design process.
- Mining contrast patterns from multivariate decision trees(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2018) Cañete Sifuentes, Leonardo Mauricio; Monroy, Raúl; emimmayorquin; Jair Escalante, Hugo; Conant Pablos, Santiago Enrique; Loyola González, Octavio; Escuela de Ingeniería y Ciencias; Campus Estado de México; Medina Pérez, Miguel AngelCurrently, there is a growing interest in the development of classifiers based on contrast patterns (CPs); this is partly due to the advantage of them being able to explain a classification result in a language that is easy to understand for an expert. Thorough experiments show that CP- based classifiers, when using contrast patterns extracted by miners based on decision trees, attain accuracies comparable with state-of-the-art classifiers like SVM, k-NN, C4.5, Bagging and Boosting. Existing decision tree-based miners use Univariate Decision Trees (UDTs) to extract CPs. For tree-based classification classifiers based on Multivariate Decision Trees (MDTs) achieve better accuracy than those based on UDTs. This result might be attributable to that MDTs use multivariate relations (e.g., 2height + 3weight > 40) which, in some cases, separate better the classes than the univariate relations (e.g., age > 40) that UDTs use. Our hypothesis runs parallel, but for CP-based classification: using CPs extracted from MDT-based miners, which we call multivariate contrast patterns, a CP-based classifier shall significantly improve on the performance of others based on UDTs. We propose an algorithm to extract, simplify and filter multivariate CPs. We make an empirical study of our proposed algorithm. We use 112 datasets, taking half of the datasets for tuning the parameters of our algorithm. To validate our hypotheses, we use the other half of the datasets as a testing set to compare our algorithm against other state-of-the-art CP miners in terms of quality, and against other state-of-the-art classifiers, in terms of classification performance. The results obtained in the testing set show that the quality of multivariate CPs, in terms of Jaccard, is significantly higher than that of CPs extracted through UDTs (univariate CPs). We also show that the classification results for CP-based classifiers are significantly better when using multivariate CPs than when using univariate CPs; which could be explained by the higher quality of multivariate CPs. The classification results for multivariate CP-based classifiers are also competitive with non-pattern-based state-of-the-art classifiers. Yet, the plus is that multivariate CP-based classifiers provide contrast patterns, which are abstract-level explanations that could help an expert to gain insights in the problem under investigation.
- Clasificación de espacios urbanos a gran escala a partir de un estudio de percepción y datos del INEGI de la ciudad de Puebla, San Pedro Cholula y San Andrés Cholula.(Instituto Tecnológico y de Estudios Superiores de Monterrey) Clavijo Plourde, Daniel; Oliart Ros, Alberto; Departamento de Ingeniería y Ciencias Computacionales; Departamento de Ingeniería y Ciencias Computacionales; Campus PueblaEl presente trabajo consiste en un estudio enfocado a la automatización de la clasificación de espacios urbanos de acuerdo a un estudio de percepción y datos sociodemográficos del INEGI. En la actualidad no se cuenta con un proceso sistematizado que facilite la toma de decisiones en relación a una planeación urbana adecuada. Por esta razón, se llevó a cabo un estudio para medir la percepción humana en 5 rubros: belleza arquitectónica, contaminación, diversión, riqueza y seguridad. La información recabada se utilizó para proponer un modelo de Machine learning que pueda realizar un reconocimiento de patrones entre la percepción obtenida y los datos demográficos. Este primer acercamiento pretende denotar los puntos clave necesarios para el desarrollo de dicho modelo y su posible implementación.