Explainable AI for Predictive Toxicology

Abstract

The amount of data and the performance of machine learning algorithms have increased dramatically as technology has advanced over the last few decades. Also in the field of toxicology, large amounts of data have been generated, and now one challenge is how this research field should deal with the available data to benefit the drug development process, including toxicity predictions. Simultaneously, drug development is getting more costly and time-consuming. The research for this thesis investigates how machine learning algorithms can improve drug development through improved predictive tools for drug toxicity.