Comorbidities are often associated with sub-optimal drug response compared with clinical trials and play an important role in providing personalized medicine therapies.

Metadvice is a startup that develops explainable AI for the management of comorbid patients, using clinical guidelines, synthetic data and electronic health records to provide feed-back to the clinicians at the point of care and to produce Real World Evidence (RWE) analytics of interest to pharma.

This presentation will show how transfer learning is used in both clinical and pharma settings with applications to blood cancers and cardio-metabolic diseases in order to identify sub-/optimal responders for targeted therapies.