Data Science

Automated Reports from Xrays

Using deep learning for automatically generated medical reports describing radiological images.

Neural networks versus Logistic regression for 30 days all-cause readmission prediction

We conclude that data from patient timelines improve 30 day readmission prediction, that a logistic regression with LASSO has equal performance to the best neural network model and that the use of administrative data result in competitive performance compared to published approaches based on richer clinical datasets.

PostDoc in Health AI

Summary The University of Zurich together with the University Hospital of Zurich are embarking on a concerted effort to develop informatics programs to advance biomedical research and healthcare using cutting edge computational approaches. As part of these efforts, the Chair of Medical Informatics (https://krauthammerlab.ch) investigates topics in clinical data science and translational bioinformatics, such as knowledge discovery from Big Data sources (Electronic Medical Records, health registries) as well as the analysis of human Omics data.

Student Project in Time Series Representation Learning

Summary Time series analysis plays an important role in various industries such as medical informatics, healthcare, financial market, and climate modeling. Recent development in machine learning has promised to revolutionize predicting and classifying time sequences, by means of temporal convolutional networks and transformers. Although many state-of-the-art models have reached good performance in time series forecasting and classification, many time series tasks remain a challenge. Our research interest lies in time series representation learning, which aims to capture both contextual and temporal information at any resolution and generate representations to improve multiple downstream tasks (patient sequence similarity, etc.

We are looking for Masters Students!

We are currently looking for Master’s students in the field of Bioinformatics for the following topics Analyzing cell-free DNA fragmentation patterns in clinical samples This project aims to improve diagnosis and prognosis through liquid biopsies in cancer and various other diseases. Cell-free DNA (cfDNA) carries information on the epigenetics of distal tissues and organs. We discern epigenetic signatures from cfDNA-sequencing data and connect those to diagnoses and patient outcomes. The student should have basic experience with Unix systems, some experience with at least one scripting language (e.