Data-driven informatics for translational and clinical research
We integrate high-dimensional healthcare data (clinical, molecular, multi-omics, imaging, and mobile health) to accelerate biomarker discovery, diagnostics, and personalised treatments. Our main areas of focus are:
Data Integration and Interoperability
We develop methods to harmonise and standardise clinical and biomedical data, ensuring secure and FAIR-compliant sharing. This includes research and infrastructure development supporting federated data analysis, ontology-based annotation, and the integration of various healthcare data sources.
Projects: ImmUniverse, HeBA, EPND, CoVaLUX.
AI and Predictive Modeling
We apply machine learning, deep learning, and statistical methods to analyse large-scale healthcare data available centrally as well as federated. Our research focuses on patient stratification, disease trajectory modelling, and development of clinical decision support tools for diagnosis, prognosis, and treatment.
Projects: IDERHA, LEOPARD.
Natural Language Processing (NLP) and Synthetic Data
We implement text mining and NLP techniques to extract insights from unstructured clinical records and biomedical literature. We also apply privacy-preserving synthetic data models to ensure data confidentiality.
Project: FAIRclinical.
Our research projects
This section introduces current projects of the Clinical and Translational Informatics group:
-
Duration:
2020-2026
-
Funding source:
European Union’s Innovative Medicines Initiative (IMI-JU)
-
Researchers:
-
Partners:
The consortium of 23 partners from various countries, including universities, research organisations, and pharmaceutical companies.
-
Description:
The ImmUniverse project aims to improve the diagnosis and treatment of immune-mediated diseases, specifically ulcerative colitis and atopic dermatitis. By exploring the interactions between immune cells and tissues at the microenvironment level, the project seeks to develop personalised therapies and better predict disease progression and treatment responses.
Data science plays a crucial role in this project by identifying disease-specific and cross-disease signatures using multi-omics approaches.
In addition to the creation of the data management plan, we are responsible of the implementation of bioinformatics approaches. More specifically we are leading the establishment of the virtual biobank platform, the sample curation process and the development and operation of data and analysis portal while ensuring data security. We are also contributing in data curation and harmonization, cross-cohort analyses and multi-omics based predication and disease classification.
-
Project details (PDF):
-
Duration:
2020-2024
-
Funding source:
European Union’s Horizon 2020 Framework Programme
-
Researchers:
-
Partners:
This project involves several internal partners within the LCSB, including Rejko Kruger, Venkata Satagopam and 26 partners from 15 countries.
-
Description:
The ORCHESTRA project aimed at addressing the COVID-19 pandemic by establishing a large-scale cohort and by focusing on generating evidence to improve prevention and treatment, with a special emphasis on at-risk populations. Key objectives include developing recommendations, assessing environmental and socio-economic impacts, evaluating vaccine efficacy, and creating models for future pandemics.
The project utilises innovative methods, such as federated learning techniques and advanced modeling capabilities, to integrate epidemiological, clinical, microbiological, and genotypic data from population-based cohorts.
We support data management and governance for the Luxembourg-based cohort within the ORCHESTRA project. Our work includes data infrastructure, engineering, and federated analysis using the DataShield-OPAL interface.
-
Project details (PDF):
-
Duration:
2021-2026
-
Funding source:
Haut-Commissariat à la protection nationale (HCPN)
-
Researchers:
-
Partners:
This project is a research initiative in Luxembourg that involves several internal partners within the LCSB, including Paul Wilmes, Alexander Skupin, Jorge Gonçalves, Venkata Satagopam, Jochen Klucken and Rejko Krüger and also the Luxembourg Institute of Health (LIH)
-
Description:
The CoVaLux project aims to address key unanswered questions related to COVID-19, particularly the impact of vaccination and the phenomenon of Long COVID. It involves an interdisciplinary approach, combining immunology, epidemiology, digital health, social science, and public health. The project builds on previous national studies and aims to provide valuable insights to inform future vaccination strategies and public health policies in Luxembourg.
The overarching strategy for data integration will be based on data analysis and subsequent integration into mechanistic epidemiological models and in a time-specific manner to reflect the distinct phases of the pandemics.
We are leading activities associated with data management, interoperability and integrated analysis of data gathered in the project using advanced machine learning methods.
-
Project details (PDF):
-
Duration:
2022-2025
-
Funding source:
Michael J Fox Foundation
-
Researchers:
-
Partners:
In Luxembourg, the HeBA project is organised under the framework of the National Centre of Excellence in Research on Parkinson’s Disease (NCER-PD) and involves Venkata Satagopam and Rejko Kruger as LCSB partners in collaboration with the Luxembourg Institute of Health (LIH) in addition to partners from 4 European countries.
-
Description:
The Healthy Brain Ageing (HeBA) project focuses on understanding and preventing age-related neurodegenerative diseases, such as Parkinson’s and Alzheimer’s. It aims to identify risk factors for developing age-related neurodegenerative diseases, develop early detection methods, and create strategies for preventing these diseases.
The project also promotes collaboration and emphasizes the importance of data sharing and open access to research findings to accelerate progress in the field.
In this project, we are involved in gathering information from the surveys, UPSIT smell test and follow-up visits. The curated data from all sites will be integrated and proceeded with federated data analysis.
-
Project details (PDF):
-
Duration:
2024-2026
-
Funding source:
CHIST-ERA, Luxembourg National Research Fund (FNR)
-
Researchers:
-
Partners:
The project involves Venkata Satagopam from LCSB, University of Luxembourg, Tim Beck from University of Nottingham, Nona Naderi from University Paris-Saclay and Patrick Ruch from the Swiss Institute of Bioinformatics.
-
Description:
The FAIRClinical proposal aims to enhance the FAIR-ness (Findable, Accessible, Interoperable, and Reusable) of supplementary data files in clinical research. The project focuses on improving the reuse of unstructured clinical case report forms (CRFs) by structuring them into the OMOP Common Data Model (CDM). This initiative seeks to make clinical data more accessible and usable for researchers, ultimately supporting better clinical research outcomes.
The project involves developing machine learning approaches for information extraction from medical and clinical research articles and their supplementary materials to enrich their contents and enhance their interoperability, findability, and reusability.
We are coordinating the project and leading the sematic structuring, relation extraction, and modelling of CRF data. We also support partners in the data collection, standardization, and the semantic enrichment.
-
Project details (PDF):
-
Duration:
2021-2026
-
Funding source:
European Union’s Innovative Medicines Initiative (IMI-JU)
-
Researchers:
-
Partners:
The project brings together a consortium of 19 partners from multidisciplinary public and private sector from 8 countries.
-
Description:
The European Platform for Neurodegenerative Diseases (EPND) project aims to accelerate the discovery of biomarkers and treatments for neurodegenerative diseases like Alzheimer’s and Parkinson’s. By creating a collaborative data platform, EPND links existing European catalogues and research infrastructures, facilitating discovery, sharing and analysis of biological samples and data.
By establishing principles for fair and transparent data governance, the project ensures the long-term sustainability for data access and reuse.
At LCSB, we are co-leading the establishment of the EPND platform. We are acting as a data hub of EPND, offering data hosting and facilitating data and sample access management. To ensure data integration, we are involved in the data characterization and harmonisation. We are also responsible for the project’s data management plan describing the data sources and processes of how this data will be exploited, shared and preserved considering ethical and legal aspects.
-
Project details (PDF):
-
Duration:
2023-2028
-
Funding source:
European Union’s Innovative Health Initiative (IHI-JU)
-
Researchers:
-
Partners:
The project brings together a consortium of 33 academic, clinical, MedTech, pharmaceutical, and IT partners, as well as patient advocacy organisations and public authorities.
-
Description:
The IDERHA (Integration of Heterogeneous Data and Evidence towards Regulatory and HTA Acceptance) project aims to integrate diverse health data to enhance patient care and medical research, focusing initially on lung cancer. Key goals include creating a federated data space, developing AI/ML algorithms for personalised disease management, implementing remote patient monitoring, and enhancing data interoperability and FAIRification framework.
In the project, we are leading the development of the IDERHA infrastructure and tools for the integration of data. That requires cross-domain federated data management, and AI/ML infrastructure for federated data analysis. We are responsible of the data management plan, and we are contributing to the long-term sustainability of data by ensuring adhesions to standard and interoperability requirements, including FAIRification of metadata and harmonization/curation of data.
-
Project details (PDF):
-
Duration:
2023-2028
-
Funding source:
European Union’s Horizon 2020 Framework Programme
-
Researchers:
-
Partners:
The project brings together a consortium of 12 partners from 8 countries.
-
Description:
The LEOPARD project (Liver Electronic Offering Platform with Artificial Intelligence-based Devices) aimed at improving liver transplantation outcomes. The project focuses on developing and validating an AI-based predictive algorithm to enhance organ allocation strategies for patients with decompensated cirrhosis (DC) and hepatocellular carcinoma (HCC).
By combining a comprehensive data integration, predictive signatures from OMICs and radiomics, the project aims to create more accurate risk assessments and prioritization tools for liver transplant candidates. This will significantly improve liver transplant outcomes, reduce waitlist mortality, and harmonize organ allocation practices across Europe.
We are leading multimodal data management and processing through the creation of a data management plan, the collection, curation, harmonization of data, the development of a sustainable and FAIR database and analysis portal for the project. In addition, we are involved in the design and implement a harmonized common up-dated pre-liver transplantation (LT) data set across participating Organ Sharing Organizations (OSOs) and of prospective cohort/clinical studies. We will also support the project in the testing and validation of predictive models and OMICs signatures.
-
Project details (PDF):