Navigation auf uzh.ch
Navigation auf uzh.ch
Harnessing the vast amounts of observational data available today requires a nuanced approach beyond mere application of machine learning techniques. Effectively decoding such multidimensional problems demands an understanding of not just patterns, but also causality and the processes underlying the data. Traditional reductionist approaches, however, often stumble in the face of high-dimensional complexity. Our aim is to marry these two methodologies, navigating the complexity while anchoring our exploration in robust scientific principles.
One such multifaceted issue is Multimorbidity (i.e., the presence of multiple chronic conditions within an individual patient) is one of the great societal health challenges of our time. The global burden is enormous and multimorbidity is the biggest driver of health care cost. High-quality, actionable research is urgently needed to inform effective, safe and patient-centered care in multimorbid patients.
Our project will develop and implement a much-needed, novel research paradigm bringing together subject knowledge, methods expertise from conventional medical research, Machine Learning (ML) and Mathematics. We seek to deliver innovative, actionable methods to transform the paltry evidence-base while appropriately handling the high degree of complexity in the treatment of patients with multiple chronic conditions. Our consortium consists of leading scientists from Epidemiology (Milo Puhan, Viktor von Wyl, Miquel Serra Burriel), Particle Physics (Nicola Serra), Pharmacoepidemiology (Andrea Burden) and Mathematics (Alessio Figalli, Ashkan Nikeghbali) as well as Medicine (Cynthia Boyd, Oliver Senn, Gregory Lucas), Health Economy (Simon Wieser) and Computer Science (Andrey Ustyuzhanin). The project is hosted the University of Zurich and the Federal Institute of Technology.
PhD position in Epidemiology:
The focus of this position will be on the development methods and applications for causal inference and benefit harm balance modelling in the area of treatments for patients with chronic conditions. Candidates should have a MSc degree in epidemiology, biostatistics, econometrics or another field that brings together data science and subject knowledge. A strong interest for quantitative methods and collaborating across fields is required.
PhD position in Pharmacoepidemiology:
Within this PhD position, the candidate will work with large population-based healthcare data to evaluate polypharmacy patterns and drug-drug interaction risks among patients with multimorbidity. The successful candidate will work on developing and comparing machine learning causal inference models to conventional epidemiologic study designs. Candidates applying for this PhD position should hold a Masters degree in pharmacoepidemiology, epidemiology, biostatistics, public health, pharmacy or another relevant field. Strong communication skills and ability to collaborate with interdisciplinary researchers is required. Additionally, prior experience with machine learning is desirable.
PhD and/or Postdoc positions at the Department of Mathematics of ETH and Department of Physics of UZH:
These positions will concentrate on the data analysis and machine learning aspects of the project. Candidates applying for the PhD position should hold a degree in Mathematics, Physics, or Computer Science and possess experience with deep learning technologies. An eagerness to delve into complex, interdisciplinary research and a passion for understanding and applying machine learning to real-world problems are essential. As for the Postdoctoral position, applicants are expected to have a robust background in the theoretical and mathematical aspects of deep learning and on the application of deep learning to the real-world data. A proven record of strong publications on the topic is essential. We are seeking individuals who are passionate about their work, have a strong commitment to research excellence, and can contribute meaningfully to our team's collective knowledge and expertise.
