Investigating ADR mechanisms with Explainable AI: a feasibility study with knowledge graph mining.

Emmanuel BressoPierre MonninCédric BousquetFrançois-Elie CalvierNdeye-Coumba NdiayeNadine PetitpainMalika Smaïl-TabboneAdrien Coulet

Published in: BMC medical informatics and decision making (2021)

Knowledge graphs provide sufficiently diverse features to enable simple and explainable models to distinguish between drugs that are causative or not for ADRs. In addition to explaining classifications, most discriminative features appear to be good candidates for investigating ADR mechanisms further.

Keyphrases

healthcare
adverse drug
artificial intelligence
convolutional neural network
machine learning
neural network