Roberto Carlos Diaz-Velazco
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Roberto Carlos Diaz-Velazco has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Alberto Gudiño-Ochoa
División de Tecnologías para la Integración Ciber-Humana, Centro Universitario de Ciencias Exactas e Ingenierías (CUCEI), Universidad de Guadalajara, Guadalajara 44430, Jalisco, Mexico
1 shared publication - Eduardo Ruiz-Velázquez
División de Tecnologías para la Integración Ciber-Humana, Centro Universitario de Ciencias Exactas e Ingenierías (CUCEI), Universidad de Guadalajara, Guadalajara 44430, Jalisco, Mexico
1 shared publication - Jorge Ivan Cuevas-Chávez
División de Tecnologías para la Integración Ciber-Humana, Centro Universitario de Ciencias Exactas e Ingenierías (CUCEI), Universidad de Guadalajara, Guadalajara 44430, Jalisco, Mexico
1 shared publication - Julio Alberto García-Rodríguez
Departamento de Eléctrica-Electrónica, Instituto Tecnológico de Ciudad Guzmán, Tecnológico Nacional de México, Ciudad Guzman 49100, Jalisco, Mexico
1 shared publication
Latest Publications
- Open Access
- 30 June 2026
- Published by MDPI AG in Technologies
- Vol. 14 (7) , 399
Foundation Time-Series Models for Local Forecasting in Adults with Type 1 Diabetes Using Continuous Glucose Monitoring Signals
Foundation time-series models have recently shown potential for forecasting complex temporal signals, but their behavior in patient-specific continuous glucose monitoring (CGM) forecasting remains insufficiently understood, particularly when only glucose history is available. This study provides a patient-level benchmark of foundation models for 30 min ahead glucose prediction in adults with type 1 diabetes mellitus (T1DM) under a strictly univariate CGM-only setting. Using the HUPA–UCM dataset from 25 individuals, we evaluated TimeGPT, Chronos, and Sundial against representative statistical, machine learning, and deep learning forecasters, including ARIMA, ETS, gradient-boosting models, recurrent networks, and neural forecasting architectures. Models were assessed using a local walk-forward validation strategy over the final 24 h of CGM data for each patient. Foundation models achieved the strongest global performance, with Sundial obtaining the lowest overall MAE (), while TimeGPT and Chronos remained among the most competitive approaches. However, patient-level analyses showed that this advantage was not uniform: ARIMA remained highly competitive in selected individuals, and no single model consistently dominated across the cohort. These findings suggest that foundation time-series models are promising tools for short-horizon CGM forecasting, but their use should be framed within patient-specific model selection rather than as universal replacements for classical forecasting methods.- 30 June 2026
- Published by MDPI AG in Technologies
- Vol. 14 (7) , 399
Open AccessFoundation Time-Series Models for Local Forecasting in Adults with Type 1 Diabetes Using Continuous Glucose Monitoring Signals
Foundation time-series models have recently shown potential for forecasting complex temporal signals, but their behavior in patient-specific continuous glucose monitoring (CGM) forecasting remains insufficiently understood, particularly when only glucose history is available. This study provides a patient-level benchmark of foundation models for 30 min ahead glucose prediction in adults with type 1 diabetes mellitus (T1DM) under a strictly univariate CGM-only setting. Using the HUPA–UCM dataset from 25 individuals, we evaluated TimeGPT, Chronos, and Sundial against representative statistical, machine learning, and deep learning forecasters, including ARIMA, ETS, gradient-boosting models, recurrent networks, and neural forecasting architectures. Models were assessed using a local walk-forward validation strategy over the final 24 h of CGM data for each patient. Foundation models achieved the strongest global performance, with Sundial obtaining the lowest overall MAE (), while TimeGPT and Chronos remained among the most competitive approaches. However, patient-level analyses showed that this advantage was not uniform: ARIMA remained highly competitive in selected individuals, and no single model consistently dominated across the cohort. These findings suggest that foundation time-series models are promising tools for short-horizon CGM forecasting, but their use should be framed within patient-specific model selection rather than as universal replacements for classical forecasting methods.