Interactive modelling and prognosis of a COVID-19 hospitalized patient via multistate models
A shiny app is presented with two main goals: 1) to fit a MSM from specific data in a friendly way (programming skills are not required); 2) to predict the clinical evolution for a given patient based on the previous MSM. For illustrative purposes, we show how the app works using data from a multico...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Fecha de publicación: | 2022 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/371945 |
| Acceso en línea: | https://hdl.handle.net/2117/371945 |
| Access Level: | acceso abierto |
| Palabra clave: | Survival analysis (Biometry) COVID-19 (Disease) Shiny app Multistate model COVID-19 Anàlisi de supervivència (Biometria) COVID-19 (Malaltia) Classificació AMS::62 Statistics::62N Survival analysis and censored data Àrees temàtiques de la UPC::Matemàtiques i estadística::Estadística matemàtica |
| Sumario: | A shiny app is presented with two main goals: 1) to fit a MSM from specific data in a friendly way (programming skills are not required); 2) to predict the clinical evolution for a given patient based on the previous MSM. For illustrative purposes, we show how the app works using data from a multicohort study of more than 5,000 hospitalized adult COVID-19 patients from 8 Catalan hospitals during the first five waves of the pandemic. Different models have been fitted for the first Catalan pandemic wave, including as states the main outcomes (discharge and death) together with objective interventions during hospitalization such as non-invasive or invasive mechanical ventilation. The application and the underlying model are intended to be very useful for clinicians and to enhance the approach in modelling the course of other diseases with different stages of severity. |
|---|