Artificial Intelligence field has seen dramatic advances in the past few years with much excitement around the use of deep learning (DL), many-layered convolutional neural networks (CNN). The world has witnessed striking advances in the ability of machines to understand and manipulate data, including images, language, and speech. CNN showed ability to detect a complex pattern in high-dimensional data, but also are able to integrate data from various resources by having many input channels into neural network. Human genetics can benefit immensely from DL.
We have previously developed deep learning framework for omics analysis GenNet. However, the application of AI in genetics analysis is still quite limited. The main issue is the restriction for data sharing between cohorts and loss of power, compare to the pooled analysis. Federated Learning is a distributed machine learning approach which enables model training on a large corpus of decentralized data.
The main goal of this project is to develop new federated learning framework for multi-center genetics analysis in close collaboration with NVIDIA company, which will be able to utilize machine learning approaches and increase power of gene discovery. We aim to apply these methods on large datasets from population-based Rotterdam study, UK Biobank as well as within world-wide genetic consortiums.
Erasmus MC is dedicated to a healthy population and excellence in healthcare. By conducting groundbreaking work, we aim to push through current boundaries and leading the way forward in research, education and healthcare. We have access to the latest equipment and techniques in a state-of-the-art environment. Our departments are known for their forefront research and facilities.
The candidate will be appointed jointly in the Departments of Radiology and Nuclear Medicine and Department of Epidemiology. You will be working in multidisciplinary team with experts in machine learning, deep learning, genomics and epidemiology. Research group has its own GPU cluster and access to the Amsterdam super-computer.
Qualifications and skills
We are seeking enthusiastic candidates with a strong motivation to engage in the development and application of advanced analytical methods, including artificial intelligence, to tackle this important societal and healthcare challenge. Successful candidates are expected to have a strong quantitative or computer science background, excel at critical thinking, and highly motivated to solve clinical and epidemiological problems.
- The candidate should have an MSc degree in Mathematics, Computer Science, Statistics, Bioinformatics, Physics, Electrical Engineering, or in an equivalent discipline.
- Experience with Python language.
- Experience with machine learning and deep learning methods.
- Experience in working with large datasets.
- Knowledge or willingness to learn more about medical imaging, epidemiology and genetics.
- Good communication and writing skills in English are required.
- Ability to work, launch and embrace collaboration in a multidisciplinary setting.
Before you apply please check our conditions for employment.
For more information about this position, please contact Dr. Gennady Roshchupkin, Assistant Professor and Computational Population Biology group leader, e-mail: firstname.lastname@example.org.
No agencies please.