Thèse de Christopher NORMAN, sous la direction de Aurélie Névéol, Patrick Bossuyt et Mariska Leeflang, dans le cadre du projet MIRoR (lien vers MIRoR)
Soutenance le 11 février à 12:00 à l'Université d'Amsterdam
Elle sera présentée en anglais devant un jury composé de :
- Patrice Bellot (Rapporteur) Aix-Marseille Université
- Maroeska M. Rovers (Rapporteur) Radboud Universiteit Nijmegen
- Nicolette F. de Keizer (Examinatrice) AMC, Universiteit van Amsterdam
- Maarten de Rijke (Examinateur) Universiteit van Amsterdam
- Sandra Bringay (Examinatrice) Université Paul-Valéry, Montpellier 3
- Allexandre Allauzen (Examinateur) Université Paris-Saclay
- Aurélie Névéol (Directeur de thèse) CNRS, LIMSI, Université Paris-Saclay
- Patrick M.M. Bossuyt (Directeur de thèse) AMC, Universiteit van Amsterdam
- Mariska M.G. Leeflang (Co-encadrant de thèse) AMC, Universiteit van Amsterdam
Recent advances in artificial intelligence have seen limited adoption in systematic reviews, and much of the systematic review process remains manual, time-consuming, and expensive. Authors conducting systematic reviews face issues throughout the systematic review process. It is difficult and time-consuming to search and retrieve, collect data, write manuscripts, and perform statistical analyses. Screening automation has been suggested as a way to reduce the workload, but uptake has been limited due to a number of issues, including licensing, steep learning curves, lack of support, and mismatches to workflow. There is a need to better align current methods to the need of the systematic review community.
Diagnostic test accuracy studies are seldom indexed in an easily retrievable way, and suffer from variable terminology and missing or inconsistently applied database labels. Methodological search queries to identify diagnostic studies therefore tend to have low accuracy, and are discouraged for use in systematic reviews. Consequently, there is a particular need for alternative methods to reduce the workload in systematic reviews of diagnostic test accuracy.
In this thesis we have explored the hypothesis that automation methods can offer an efficient way to make the systematic review process quicker and less expensive, provided we can identify and overcome barriers to their adoption. Automated methods have the opportunity to make the process cheaper as well as more transparent, accountable, and reproducible.
