Bayesian Network structure learning algorithm for highly missing and non imputable data: Application to breast cancer radiotherapy data.
Fiche publication
Date publication
janvier 2024
Journal
Artificial intelligence in medicine
Auteurs
Membres identifiés du Cancéropôle Est :
Dr CLAVIER Jean-Baptiste, Dr GUIHARD Sébastien
Tous les auteurs :
Piot M, Bertrand F, Guihard S, Clavier JB, Maumy M
Lien Pubmed
Résumé
It is not uncommon for real-life data produced in healthcare to have a higher proportion of missing data than in other scopes. To take into account these missing data, imputation is not always desired by healthcare experts, and complete case analysis can lead to a significant loss of data. The algorithm proposed here, allows the learning of Bayesian Network graphs when both imputation and complete case analysis are not possible. The learning process is based on a set of local bootstrap learnings performed on complete sub-datasets which are then aggregated and locally optimized. This learning method presents competitive results compared to other structure learning algorithms, whatever the mechanism of missing data.
Mots clés
AI for healthcare, Bayesian Networks, Breast cancer, Electronic health records, Missing data, Structural learning
Référence
Artif Intell Med. 2024 01;147:102743