Applying Joint Modelling Regression Approaches in Biomedical Data Science

In recent years, the technological revolution is allowing the collection of an enormous amount of data of different types, creating enormously complex databases that require the collaboration of statisticians and clinicians to carry out a biomedical study with guarantees, applying the tools of data...

Descripción completa

Detalles Bibliográficos
Autor: Díaz Louzao, Carla
Tipo de recurso: tesis doctoral
Fecha de publicación:2022
País:España
Institución:Universidad de Santiago de Compostela (USC)
Repositorio:Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Idioma:inglés
OAI Identifier:oai:minerva.usc.gal:10347/29902
Acceso en línea:http://hdl.handle.net/10347/29902
Access Level:acceso abierto
Palabra clave:Materias::Investigación::32 Ciencias médicas::3212 Salud pública
Materias::Investigación::32 Ciencias médicas::3202 Epidemologia
Materias::Investigación::12 Matemáticas::1209 Estadística::120909 Análisis multivariante
Descripción
Sumario:In recent years, the technological revolution is allowing the collection of an enormous amount of data of different types, creating enormously complex databases that require the collaboration of statisticians and clinicians to carry out a biomedical study with guarantees, applying the tools of data science. This requires the development of new statistical techniques. This thesis focuses on joint modelling regression models for multivariate responses. Specifically, we study the cases of two and three continuous outcomes, as well as models for longitudinal and survival data. These techniques are applied in three studies of major epidemiological importance: liver damage and survival in COVID-19 patients, perinatal mental health during the COVID-19 pandemic, and the study of thyroid-related hormones.