Machine learning based radiomics-genomics integration for advanced disease quantification
POSITION DESCRIPTION
-Research Project / Research Group Description:
The student will join the euCanSHare H2020 project (6 million Euros) funded by the European Commission (2018-2022) and coordinated by Dr. Karim Lekadir. This project and its 16 members from Europe and Canada will build a big data platform for cardiovascular research that will integrate more than 35 cohorts totaling over one million records. As part of the project, a multi-omics and multi-factorial data analytics platform will be developed to exploit the large and heterogeneous datasets, to extract new biomedical knowledge, and to derive new predictive models of disease risk with statistical power nevwer achieved before.
-Job position description:
The student will focus on the development and application of computational methods, from data integration to statistical analysis and machine learning, ultimately to learn patterns and models from large-scale imaging and genetic data. With these methods, the predictive models of disease risk will integrate patient-specific information from the genetic and molecular scales up to the tissue and organ levels, thus providing a more complete picture of health and disease. Specifically, the student will build disease- and group-specific radiomic-genomic atlases and correlation maps that will inform on relevant associations and clusters between imaging phenotypes and specific genes. Subsequently, (s)he will estimate new quantitative biomarkers integrating imaging, clinical and genetic data for personalised disease quantification. Finally, the student will demonstrate the clinical potential of these models for early risk assessment of cardiovascular events (e.g. myocardial infarction, stroke) based on big data from the euCanSHare project (> 100,000s cases). For this work, the student will focus on advanced machine learning techniques such as deep learning, kernel methods and network analysis. The candidate should have an MEng/MSc degree in engineering, mathematics, computer science or other relevant field, excellent computer programming skills (C++, Matlab, and/or Python), and preferentially experience in image processing and machine learning techniques. Due to the collaboration within the euCanSHare project with collaborators in Europe and Canada, advanced oral and writing English knowledge are required.
Contact person: [email protected]