Condition assessment of patients with Type 1 diabetes using compositional data analysis

Abstract

Type 1 Diabetes Mellitus (T1DM) is a chronic disease that leads to absolute insulin deficiency. People with T1DM need exogenous insulin to maintain glucose at proper levels. Achieving optimal glycemic control is a major hurdle due to the large intra-patient variability, and continuous glucose monitoring (CGM) plays an essential role for individuals with T1DM. This thesis is devoted to describe the condition assessment of patients with T1DM through the analysis of glucose data obtained from CGM. This work focuses on understanding and dissecting the CGM measures. A model of the error of a CGM sensor has been obtained and the accuracy of the CGM has been assessed during challenging conditions. Additionally, a novel approach for the categorization of daily glucose profiles based on the analysis of compositional data (CoDa) is proposed. Finally, a probabilistic model of transition between different categories of periods of glucose data obtained with CoDa analysis is presented.

La diabetes mellitus tipo 1 (T1DM) es una enfermedad crónica que conduce a una deficiencia absoluta de insulina. Las personas con T1DM requieren insulina exógena para mantener los niveles de glucosa apropiados. Alcanzar un control glicémico óptimo tiene una gran dificultad debido a la variabilidad intrapaciente, y el monitoreo continuo de glucosa (CGM) desempeña un papel esencial para los individuos con T1DM. Este trabajo se enfoca en entender y diseccionar las medidas obtenidas de CGM. Se ha obtenido un modelo de error de un sensor CGM y se ha evaluado la precisión del CGM en condiciones difíciles. Además, se presenta un nuevo enfoque para la caracterización de perfiles de glucosa diarios con base en el análisis de datos composicionales (CoDa). Finalmente, se presenta un modelo de transición probabilístico entre diferentes categorías de periodos de datos de glucosa que fue obtenido usando técnicas CoDa.