BD.20.004 – TARES: Technology Assisted Review and Evidence Synthesis
Any decision making in health or preventive care, in any medical domain, is best served with a rapid, comprehensive and unbiased overview of the available data and evidence. Synthesis of all current data to high-quality evidence is worldwide seen as the gold standard method to achieve this. All national and international guideline organisations rely in their clinical guidance and guidelines on data and evidence synthesis reports. However, the current largely manual process of evidence synthesis is unsustainable in the era of big and diverse data and 4 million articles per year in the biomedical literature addressing medical interventions or innovations. Artificial intelligence (AI) is often seen as the solution, but yet hardly used for this purpose – certainly not to support the entire data and evidence synthesis process. The challenge lies in developing techniques that function in the highly sensitive domain of medicine where high standards and transparency are mandatory due to the impact and consequences of decisions. We have established a large international consortium, including three Dutch Universities, several medical guideline organisations, healthcare policy organisations and private partners, with leading experts in the field. Using a multidisciplinary research approach, we combine expertise from AI, such as machine learning, natural language processing, and knowledge representation with clinical research, statistics and health policy making. We will address all steps in the evidence synthesis process- from evidence retrieval, updating, quality assessment, analytical synthesis to reporting. Our developed technology will help patients, the general public, healthcare professionals, guideline and policy makers, and private companies in the field to enhance high quality rapid evidence synthesis and subsequent knowledge translation using novel AI based techniques across the entire synthesis process. We thus provide enhanced support for priority setting, health policy decision making, and shared-decision making.
Persons involved: Prof. Dr. Maarten de Rijke, Prof. Dr. Evangelos Kanoulas, Prof Dr. Paul Groth (University of Amsterdam); Prof. Dr. Frank van Harmelen, Dr. Annette ten Tije, Prof. Dr. Piek Vossen (Vrije Universiteit), Dr Lotty Hooft (Cochrane Nederland), Rene Spijker(Amsterdam UMC, Cochrane Nederland), Dr. Stefan Leijnen (HU), Dr. Marc Teunis (HU), Ir Teus van Barneveld (Kennis instituut van de Federatie Medisch specialisten), Dr. J de Jong (NHG), Heleen Post (PFN), Mark Sheehan (Elsevier)
Artificial Intelligence, Big and Diverse Data, clinical guidelines, Evidence Synthesis, Knowledge engineering, Linked data, machine learning, Natural Language Processing, Research waste, Responsible Research and Data use, semantic web
Amsterdam UMC, Cochrane Nederland, HU, Kennis instituut van de Federatie Medisch specialisten, Mark Sheehan, NHG, PFN, University of Amsterdam (UvA), Vrije Universiteit Amsterdam (VU)
|Organisation||Julius Center for Health Sciences and Primary Care; University Medical Center Utrecht (UMCU); University of Utrecht (UU)|
|Name||Prof. dr. K.G.M. (Carl) Moons|