Journal of diabetes science and technology · 2025
Probabilistic Meal Detection and Estimation in Type 1 Diabetes With Extreme Shape Variability.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Sontakke Mrunal, Cameron Faye, Bequette B Wayne
Аннотация
BACKGROUND: The study conceptualizes the application of particle filters for meal detection and shape estimation, aimed at assisting patients with type 1 diabetes (T1D) in meal announcements. It concentrates on leveraging continuous glucose monitoring (CGM) and insulin pump data to detect meals, which will eventually enhance prior research that relied on accelerometer data for event detection.
METHOD: The research employs glucose appearance rate (RA) curves derived from 21 triple-tracer studies. These curves are normalized and adjusted based on population-level meal size distributions obtained from the National Health and Nutrition Examination Survey (NHANES). Furthermore, glucose response curves resulting from insulin, as reported in existing literature focusing on fast-acting insulin analogs, are integrated into the analysis. This normalization of the population level probability distribution facilitates individualized scaling, taking into account insulin sensitivity and carbohydrate-to-insulin ratios.
RESULTS: Preliminary findings from the Tidepool data set suggest that the algorithm is effective in meal detection.
CONCLUSIONS: This innovative approach holds the potential to help individuals with T1D better manage their blood glucose levels by providing information regarding glucose response to meals and estimated meal sizes for closed-loop control. Future research can focus on enhancing key components of the algorithm and incorporating additional data types to improve its performance further.
METHOD: The research employs glucose appearance rate (RA) curves derived from 21 triple-tracer studies. These curves are normalized and adjusted based on population-level meal size distributions obtained from the National Health and Nutrition Examination Survey (NHANES). Furthermore, glucose response curves resulting from insulin, as reported in existing literature focusing on fast-acting insulin analogs, are integrated into the analysis. This normalization of the population level probability distribution facilitates individualized scaling, taking into account insulin sensitivity and carbohydrate-to-insulin ratios.
RESULTS: Preliminary findings from the Tidepool data set suggest that the algorithm is effective in meal detection.
CONCLUSIONS: This innovative approach holds the potential to help individuals with T1D better manage their blood glucose levels by providing information regarding glucose response to meals and estimated meal sizes for closed-loop control. Future research can focus on enhancing key components of the algorithm and incorporating additional data types to improve its performance further.