Fakultäten » Medizinische Fakultät » Infektionskrankheiten und Spitalhygiene, Klinik für » Prof. Dr. Huldrych Günthard » Beerenwinkel

Completed research project

Title / Titel Dealing with data sparseness of combination therapies in the SHCS dataset
PDF Abstract (PDF, 14 KB)
Summary / Zusammenfassung Background:

Finding an optimal anti-HIV treatment for a given patient is very challenging due to the large amount of information that needs to be considered. Applying advanced statistical learning methodologies on the large amount of available clinical data offers a framework for an automated approach for HIV therapy screening that can assist this process.

Study Aims:

The main goal of this study is to analyze the treatment change episodes (TCEs) available in the SHSC database using modern statistical learning methods and thereby enhance the clinical management of HIV patients.

Study Design:

First, machine learning models for HIV therapy screening that deal with the sparse, uneven therapy representation are applied on the relevant samples from the SHCS dataset (extracted using the quality control filters derived in a previous SHCS project #629). Next, we model viral evolutionary escape from the selective pressure of each antiretroviral combination therapy used in clinical practice.
Project leadership and contacts /
Projektleitung und Kontakte
Prof. Niko Beerenwinkel (Project Leader)
Other links to external web pages
Funding source(s) /
Unterstützt durch
SNF (Personen- und Projektförderung)
SNF, Swiss HIV Cohort Study
In collaboration with /
In Zusammenarbeit mit
Prof. Dr. med. H. Günthard, USZ,IMV,UZH, Zürich
Dr. J. Bogojewska, IBM, Zürich
The Swiss HIV Cohort Study
The Zurich Primary HIV Infection Study
Duration of Project / Projektdauer Jan 2015 to Dec 2017