Jump to contentJump to search

Explainable markers for cancer drug screening data

Battling cancer is one of the biggest challenges in medicine. In collaboration with the university clinic we use drug screening and gene expression data to build explainable and interpretable machine learning models, which - in contrast to less explainable/interpretable machine learning models - apply Boolean logic. We envision that our method can not only be used for finding appropriate drugs for cancer treatment, but also for gaining more insight into the complex processes of cancer. Furthermore, other applications like classifying cancer types are also conceivable.

Responsible for the content: