Postdoctoral fellow (Research Associate) in Federated Learning and Analysis

at  Universit du Luxembourg

The Luxembourg Centre for Systems Biomedicine (LCSB) is an interdisciplinary research centre of the University of Luxembourg. We conduct fundamental and translational research in the field of Systems Biology and Biomedicine – in the lab, in the clinic and in silico. We focus on neurodegeneration and are especially interested in Alzheimer’s and Parkinson’s disease and their contributing factors.
The LCSB recruits talented scientists from various disciplines. Computer scientists, mathematicians, biologists, chemists, engineers, physicists and clinicians from more than 50 countries currently work at the LCSB. We excel because we are truly interdisciplinary and together, we contribute to science and society.
Successful candidates will join the Bioinformatics Core, led by Prof. Reinhard Schneider, which focuses on managing and analysing complex biomedical and clinical data. We are internationally recognized for our GDPR compliant data hosting solutions and data management systems. We develop innovative methodologies for data mining, federated data analysis, and data FAIRification. We strongly advocate following responsible and reproducible research (R3) principles and best practices in software development. For more information, please visit our website.
is integral part of the CLINNOVA project, an international initiative of leading clinicians and scientists from university hospitals, private clinics, and health research institutes across Luxembourg, France, Germany, and Switzerland. The project aims to revolutionize healthcare by harnessing the power of data federation, standardization, and interoperability to advance precision medicine for treatment decisions. To learn more about the CLINNOVA project and its objectives, visit: https://www.uni.lu/fr/news/clinnova-to-launch-precision-medicine-initiative-across-europe/

REQUIRED QUALIFICATIONS:

• A PhD in computer science, information technology, computational biology, bioinformatics, or a related field, with keen interest in health research and related IT infrastructure
• Domain knowledge:
• Good understanding of statistical analysis principles and AI/ML techniques in both centralized and federated environments
• Hands-on experience in developing, deploying, and maintaining ML operations (MLOps) within IT infrastructure, including familiarity with virtualization and containerization technologies such as Docker and Kubernetes is considered advantageous
• Technical skills:
• Proficiency in Python programming language, including data manipulation libraries (e.g., Pandas), ML frameworks (e.g., scikit-learn), and visualization tools (e.g., Matplotlib, Seaborn)
• Familiarity with federated technologies, such as Flower, NVIDIA FLARE, is considered advantageous
• Additional skills:
• Good ability to manage tasks effectively to meet project deadlines and reporting
• Experience with authentication and authorization solutions (e.g., ELIXIR-AAI, OIDC, GA4GH), cloud-based AI/ML services, testing methodologies (unit, integration, e2e) is considered advantageous

As a Postdoctoral Researcher, you will develop and implement federated analytical workflows tailored for health research. You will apply AI/ML algorithms to analyse a diverse range of data types, including clinical, molecular (-omics), and (sensor/mobile and PROMs/PREMs) within a federated environment. Additionally, you will innovate state-of-the-art federated AI/ML methods, to ensure privacy and data security in clinical research. To augment federated analysis, you will be generating synthetic data using ML techniques, such as Generative Adversarial Networks (GANs). Your workflows and methods will be incorporated by a multidisciplinary team into the CLINNOVA platform for federated data management and analysis. You will take an active role on project activities and effectively disseminating findings to the project members and the scientific community through project meeting, conferences and publications. Your main tasks will include:

• Develop federated analytical workflows: integrate and adapt federated learning workflows specifically designed for health research, emphasizing scalability, efficiency, and privacy
• Analyse diverse data sets with AI/ML: apply advanced AI/ML algorithms to a broad spectrum of health data, including clinical, molecular (-omics), and real-world data, in a federated context
• Generate and use synthetic data: create synthetic data using methods like GANs for use in federated analysis, ensuring the data is both realistic and privacy compliant
• Support the development of federated data management and analysis platform: actively engage and support the platform development team in implementing federated analytical workflows into the CLINNOVA platform
• Take an active role on project activities: take charge of specific project activities, working in close harmony with a multidisciplinary team to meet project goals effectively
• Disseminate research findings: actively share the ongoing work and findings with the project members, the scientific community and other stakeholders through project meetings, conferences and publications