Sirvi Autor "Soodla, Sander" järgi

Nüüd näidatakse 1 - 1 1

R-pakett diagnooside vaheliste seoste uurimiseks elukestusanalüüsiga
(Tartu Ülikool, 2025) Soodla, Sander; Ilves, Neeme, juhendaja; Malk, Maria, juhendaja; Kolde, Raivo, juhendaja; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Arvutiteaduse instituut
The aim of this bachelor's thesis was to create an R software package for investigating effects between medical diagnoses using survival analysis, i.e., statistical methods that analyze the time until the occurrence of a specific event. The developed package, Exposure2OutcomeSurv, works with health databases in the OMOP Common Data Model (CDM) format. The package enables health informatics researchers to investigate whether the occurrence of one diagnosis (exposure) affects the occurrence of another diagnosis (outcome), simultaneously for many different diagnoses. For this purpose, it compares the time to outcome occurrence between patients with the exposure and a control group, matched by gender and age but without the exposure. The package visualizes the probability of outcome occurrence over time, statistically compares the groups, and estimates the magnitude of the outcome risk in one group compared to the other. To enhance user-friendliness, the package includes a Shiny graphical user interface. The package's functionality was demonstrated with a pilot study.