Our mission is that one day every single person gets the treatment they deserve in a timely manner. Our healthcare system of today is not good enough. One of the main challenges of today’s healthcare is that disease management is mainly focusing on insights gathered at population level. We believe that a “next-generation of management” can be achieved when we supercharge with insights gained from Big Data. We are convinced that Data Saves Lives. Therefore, we investigate new methods to handle and analyse Big Data, with a specific focus on the chronic disease multiple sclerosis (MS).
Multiple Sclerosis (MS) is a progressive demyelinating and degenerative immune-mediated disorder of the central nervous system with symptoms depending on the disease type and the site of lesions. The disease course is unpredictable and heterogeneous. Not only physical (e.g. visual and cognitive function), but psychological and social aspects as well are affected in patients with MS. Therefore, MS should be featured by an individualized and intense clinical follow-up and multidisciplinary treatment.
High performance MS-specific decision support systems are needed to support treatment decision-making by neurologists and regulators (= the right disease modifying therapy (DMT) for the right patient). And we need these decision support systems to function well in what we call “a real-world” setting. To date, 14 DMTs have been approved for relapsing-remitting MS on the basis of their efficacy in randomized controlled trials (RCTs). RCTs are accepted as the gold standard for assessing the efficacy and safety of any new drug, and are conducted in a controlled setting with well-defined homogeneous patient populations selected through strict inclusion criteria. These cohorts do not necessarily represent MS in real life and conclusions made from these RCTs therefore do not always translate to the individual patient
Real-world data (RWD) are defined as data derived from a number of sources that are associated with outcomes in a heterogeneous patient population representing the real-world settings. To transform the care of PwMS, we need to speed-up diagnosis, prognosis and treatment. We are convinced that the key to accomplish this is to develop and implement new methodologies to handle and analyse real-world MS datasets.
Our research in the upcoming years focusses on following 3 clinical challenges:
- Identification of new biomarkers for disease activity and – progression
- Develop decision-support systems for relative treatment effectiveness in a real-world setting
- Develop tools and methodologies to support scaling-up real-world MS data research
Thijs Becker, Post-doc:
Thijs got his PhD in theoretical physics in 2015 from Hasselt University. Afterwards, he started a postdoc in applied machine learning in the group of prof. Christian Van den Broeck. He is currently working on models that predict MS disease progression, with a focus on finding new features in evoked potentials and on handling complex trajectory data. His other research interests include trustworthy machine learning (well-calibrated models, classification with a deferral option), and interpretable machine learning.
Lotte Geys, Post-doc/Project manager:
Lotte got her PhD in Biomedical sciences in 2017 and started working as clinical data manager afterwards. In December 2019, she found her way back to the academic world and started working for Liesbet Peeters as project manager. She is Liesbet’s right-hand woman, and is involved in follow-up of nearly all the projects and activities of the group, including the MSDA activities.
Hamza Khan, PhD Student:
After working as a dental surgeon for one year, Hamza got his Master degree in Public Health (MPH) in 2019. Realising the importance of data in the healthcare domain and the need for acquaintance with robust statistical methods/machine learning during his internship at the migrant health division (MHD) of the International Organization for Migration (IOM - Regional Office for EU/EEA and NATO, Brussels), he enrolled in another master degree program at Tilburg University, The Netherlands. He graduated with a Master in Cognitive Science and Artificial Intelligence (MCSAI) in August of 2020 after which he started his interdisciplinary PhD at BioMed UHasselt and Department of Precision Medicine at UMaastricht.
The focus of his PhD is finding biomarkers for the diagnosis of multiple sclerosis by applying AI/Machine learning algorithms on evoked potential time series data and MRIs. The other areas of focus of his research include improvement of the quality of life and personalized treatment plan for MS patients.
Tina Parciak, PhD Student:
Tina got her Master’s degree in Informatics with focus on Medical Informatics in 2013. Since 2014, she has been working at the Department of Medical Informatics of the University Medical Center Göttingen, focussing on IT infrastructure and data processing in the field of multiple sclerosis. Tina specialises on the topic of data harmonisation where she first was introduced to in the EUReMS project, where European MS registries collaborated to estimate the feasibility of joining the distributed and heterogeneous registry data for joint analyses. Since 2019 Tina is involved in the Multiple Sclerosis Data Alliance, where she has been responsible for the data harmonisation aspect within the FAIR spectrum. That partnership resulted in a joint PhD position with the BIOMED institute of University Hasselt, with Liesbet Peeters as promoter. In the PhD she focuses on the automatisation of data harmonisation for real-world MS data.
Ashkan Pirmani, PhD Student:
Ashkan got his master degree in industrial engineering in 2018 and after that did another master's in MBA. He has started his adventure as a joint Ph.D. student in mid-2019 with the close collaboration of STADUIS/ESAT lab of KU Leuven.
He is currently working on a federated learning framework to apply this framework in multiple sclerosis using real-world data. He believes in FAIR data. As "Accessibility" is one of these FAIR data principles, He aims to launch infrastructure and framework that learns from other data sources without sacrificing privacy in real-world data. He believes we already pass the centralized data silos era for big data research, and we have to come up with ideas to know how to share data, of course, by taking into account all the ethical and legal challenges.
His other research interest is machine learning, specifically artificial neural networks, BlockChain, and DevOps, with a keen interest in containerization.
Jan Yperman, Post-doc
Jan got his master’s degree at KU Leuven in 2015, after which he started a PhD on applied machine learning in the group of prof. Christian Van den Broeck. During his PhD he has worked on various machine learning projects, ranging from automated traffic analysis to healthcare decision support systems. The main topic of the PhD involved the gathering and analysis of Motor Evoked Potentials (MEP) in MS using machine learning. He finished his PhD in computer science in 2021.
Edward De Brouwer, PhD Student (KULeuven):
Edward graduated in Electrical Engineering from KU Leuven and UCLouvain in 2014 and started his PhD at KU Leuven in 2017 under the supervision of Prof. Yves Moreau. His main interests lie in machine learning algorithms for dealing with clinical patient trajectories. He’s been collaborating on the MS initiative since the beginning of his PhD, working on predicting disability progression worsening using patient clinical history and in the COVID-19 initiative.