Leveraging Real-World Data to optimise pharmacotherapy outcomes in multimorbid patients by using machine learning and knowledge representation methods.
LEAPfROG aims for a scientific breakthrough in Artificial Intelligence (AI) application in healthcare by combining machine learning and knowledge representation methods to deliver novel AI-powered tools, methods, models, and a prototype for AI-based decision support system. All to support effective and safe pharmacotherapy in people with multimorbidity. The value of LEAPfROG’s approach will be demonstrated via a clinically relevant and urgent use case of drug-induced kidney diseases (DIKD) in patients with chronic kidney disease (CKD).
The LEAPfROG project consists of four scientific work packages (WP).
Project Leader: dr. Joanna Klopotowska
WP1: Real-world data quality optimization
WP Team: Dr. Ronald Cornet (WP Leader), Dr. Iacopo Vagliano, Prof. dr. Ron Herings, Dr. Joanna Klopotowska, Dr. Birgit Damoiseaux,
Pharmo Institute, CASTOR, Digital Health Link
The main goal of this WP is to address problems related to limited Electronic Health Record (EHR) data quality. To overcome these problems, an infrastructure is needed tin which EHR data quality can be assessed, and data
is provided following the Findable, Accessible, Interoperable, Reusable (FAIR) principles. Establishing and populating such infrastructure from raw EHR data is currently a resource intensive manual process. Therefore, this
WP aims to contribute to minimizing the effort of establishing high-quality, enriched, and FAIR data that is needed to study pharmacotherapy outcomes in multimorbid patients.
Key elements: phenotyping, FAIRification, ontologies, semantic enrichment, heuristics, real-world data ethics
WP2: Knowledge Representation
WP Team: Prof. Annette ten Teije (WP Leader), Prof. dr. Frank van Harmelen, Romy Vos (PhD Student), Prof. dr. Cornelis Boersma, Dr. Iacopo Vagliano, Dr. Joanna Klopotowska,
The main goal of this WP is to address the problems related to lack of computer-interpretable domain knowledge needed to improve logical reasoning of machine learning, as well as lack of decision-support for drug-side effects
needed to quickly identify and compare drug-side effects profiles. Therefore, this WP aims to deliver a computer-interpretable domain knowledge in the context of DIKD that is suitable to support humans and machines. This
computer-interpretable DIKD knowledge will be in both symbolic, sub-symbolic and hybrid forms. Combinations of symbolic and sub-symbolic information are currently a technical challenge in modern AI.
Key elements: (interlinked) knowledge graphs, symbolic and sub-symbolic information and representation
WP3: Machine Learning and reasoning
WP Team: Dr. Giovanni Cina (WP Leader),
Prof. dr. Ameen Abu-Hanna,
Prof. dr. Frank van Harmelen,
Dr. Iacopo Vagliano, Dr. Joanna Klopotowska
The goal of this WP is to address problems related to the black-box nature of machine learning models and lack of logical reasoning. This WP aims to deliver explainable machine learning models for risk prediction and causal
inference in the context of DIKD in people with CKD. The models delivered need to provide actionable advises to support doctors and pharmacist in safe and effective drug prescribing.
Key elements: machine learning, explainable AI, risk prediction, causal inference, AI ethics
WP4: Human-computer collaboration
WP Team: Dr. Stephanie Medlock (WP Leader), Dr. Linda Dusseljee-Peute,
Prof. dr. Ameen Abu-Hanna, InsightRX
The practical goal of this WP is to allow the results from WP1, WP2 and WP3 to be applied in clinical practice by integrating them in LEAPfROG AI-assistant “LEA”. The scientific goal is to gain insight into shared-decision
making process surrounding DIKD and learn how LEA can improve human-computer collaboration by being of best assistance for patients and physicians. In this WP information needs of the end-users as well as technical needs
to integrate LEA in EHR will be investigated.
Key elements: clinical workflow and information needs, user experience design, EHR integration
LEAPfROG is a collaboration between the departments of Medical Informatics of AMC, Computer Science of VU; Epidemiology & Data Science of VUmc; Management Sciences of Open University; Ethics, Law & Humanities of VUmc, seven private companies; and various professional and public organisations.
The project started on the 1st of November 2022 and has a duration of 5 years.
Ethics Guidance throughout the project will be provided by:
Prof. dr. Mariette van den Hoven
The LEAPfROG project is funded by the Dutch Research Council (NWO) within the program “Knowledge and Innovation Agenda (KIA) – Key Technologies 2020”. Key technologies are technologies of the “future” that will have an enormous impact on how we live, learn, work and produce. Digital Technologies constitute such key technologies and encompass among other AI, big data and data analytics that are the focus of LEAPfROG.
LEAPfROG is co-funded by: Nierstichting, Stichting NICE, Pharmo Institute, Castor,
Z-Index, Digital Health Link and InsightRX
In the LEAPfROG project we will leverage three real-world data sources: 1) data from Electronic Health Record (EHR) system of Amsterdam UMC, 2) data from Academisch Netwerk Huisartsen Amsterdam, and 3) data from Pharmo Database Network.
Together these three real-world data sources contain information about +/- 165.000 patients with CKD.