Browsing by Author "Langarica, Saul"

Now showing 1 - 5 of 5
  • No Thumbnail Available
    Item
    A meta-learning approach to personalized blood glucose prediction in type 1 diabetes
    (2023) Langarica, Saul; Rodriguez-Fernandez, Maria; Nunez, Felipe; Doyle III, Francis J.
    Accurate blood glucose prediction is a critical element in modern artificial pancreas systems. Recently, many deep learning-based models have been proposed for glucose prediction, showing encouraging results in population modeling. However, due to the large amount of data required for training deep learning -based models, few studies have successfully addressed personalized modeling, which is critical to ensure safe policies in a closed-loop scheme given the high inter-patient variability. To address this concern, we propose a meta-learning-based technique for accurate personalized modeling that requires minimal data volume to personalize from its population version, needs few training iterations, and has a low risk of over-fitting. Results using the UVA/Padova simulator show that the proposed technique generalizes better and outperforms other approaches in standard and task-specific metrics, particularly for longer prediction horizons and higher degrees of distributional shifts.
  • No Thumbnail Available
    Item
    A Probabilistic Approach to Blood Glucose Prediction in Type 1 Diabetes Under Meal Uncertainties
    (2023) Langarica, Saul; Rodriguez-Fernandez, Maria; Doyle III, Francis J.; Nunez, Felipe
    Currently, most reliable and commercialized artificial pancreas systems for type 1 diabetes are hybrid closed-loop systems, which require the user to announce every meal and its size. However, estimating the amount of carbohydrates in a meal and announcing each and every meal is an error-prone process that introduces important uncertainties to the problem, which when not considered, lead to sub-optimal outcomes of the controller. To address this problem, we propose a novel deep-learning-based model for probabilistic glucose prediction, called the Input and State Recurrent Kalman Network (ISRKN), which consists in the incorporation of an input and state Kalman filter in the latent space of a deep neural network so that the posterior distributions can be computed in closed form and the uncertainty can be propagated using the Kalman equations. In addition, the proposed architecture allows explicit estimation of the meal uncertainty distribution, whose parameters are encoded in the filter parameters. Results using the UVA/Padova simulator and data from a clinical trial show that the proposed model outperforms other probabilistic models using several probabilistic metrics across different degrees of distributional shifts.
  • No Thumbnail Available
    Item
    Contrastive blind denoising autoencoder for real time denoising of industrial IoT sensor data
    (2023) Langarica, Saul; Nunez, Felipe
    In an industrial IoT setting, ensuring the quality of sensor data is a must when data-driven algorithms operate on the upper layers of the control system. Unfortunately, the common place in industrial facilities is to find sensor time series heavily corrupted by noise and outliers. This work proposes a purely data-driven self-supervised learning-based approach, based on a blind denoising autoencoder, for real time denoising of industrial sensor data. The term blind stresses that no prior knowledge about the noise is required for denoising, in contrast to typical denoising autoencoders. Blind denoising is achieved by using a noise contrastive estimation (NCE) regularization on the latent space of the autoencoder, which not only helps to denoise but also induces a meaningful and smooth latent space that can be exploited in other downstream tasks. Experimental evaluation in both a simulated system and a real industrial process shows that the proposed technique outperforms classical denoising methods.
  • No Thumbnail Available
    Item
    Deep Learning-Based Glucose Prediction Models: A Guide for Practitioners and a Curated Dataset for Improved Diabetes Management
    (2024) Langarica, Saul; De La Vega, Diego; Cariman, Nawel; Miranda, Martin; Andrade, David C.; Nunez, Felipe; Rodriguez-Fernandez, Maria
    Accurate short- and mid-term blood glucose predictions are crucial for patients with diabetes struggling to maintain healthy glucose levels, as well as for individuals at risk of developing the disease. Consequently, numerous efforts from the scientific community have focused on developing predictive models for glucose levels. This study harnesses physiological data collected from wearable sensors to construct a series of data-driven models based on deep learning approaches. We systematically compare these models to offer insights for practitioners and researchers venturing into glucose prediction using deep learning techniques. Key questions addressed in this work encompass the comparison of various deep learning architectures for this task, determining the optimal set of input variables for accurate glucose prediction, comparing population-wide, fine-tuned, and personalized models, and assessing the impact of an individual's data volume on model performance. Additionally, as part of our outcomes, we introduce a meticulously curated dataset inclusive of data from both healthy individuals and those with diabetes, recorded in free-living conditions. This dataset aims to foster research in this domain and facilitate equitable comparisons among researchers.
  • No Thumbnail Available
    Item
    Early-Warning System for Supervision of Urban Water Services: Case Study of Coquimbo, Chile
    (2023) Aguirre, Paula; Bravo, Marilyn; Torres, Mario; Langarica, Saul; Oyarzun, Muriel; Nunez, Felipe
    The combination of rapid urbanization, population growth, and the hydric stress due to climate change effects demand innovative, optimized approaches to the operation and supervision of urban water services. In Chile, the Superintendency of Sanitary Services has underscored the need for automatized data integration and analysis tools that foster an evidence-based, preventive approach to the supervision of urban water systems. This motivates the development of a pilot supervision and early-warning system, conceived as a cybernetic entity whose objective is to enable efficient access, analysis, and predictive modeling of the data provided by water service companies, so as to identify risks and inefficiencies in water services, monitor their evolution, and anticipate possible failures. This article discusses the development and implementation of a prototype system that provides tools for visualization, statistical and temporal analysis of georeferenced data on water pressures, network disruptions, and client complaints and deploys machine learning capabilities for predicting the quality of service indicators at different locations. The initial implementation in one region of Chile has been shown to expedite the exploitation of data on urban water services, reduce time lags in the detection of service disruptions, and generate evidence for the planning and execution of supervision activities. Based on this successful pilot, a roadmap for geographical and technological expansion is formulated, including other technological, organizational, and regulatory gaps that must be addressed to establish a data-driven framework for the supervision of urban water services.