In this talk, you will learn how we create detailed insights from different sources of medical data at Diagnostic Robotics. This consequently provides great opportunities for AI in healthcare. We will discuss the NLP challenges of working with medical reports in Hebrew and present our auto-tagging ML model for automated entities extraction from medical summaries. Our pipeline includes novelty deep model architectures built from scratch for sentence splitting, negation detection, entity relations, and term expansions. In addition, we will discuss the challenges of working with claims data, a form of health-related administrative data, to build predictive and proactive models. The talk will also cover the concept of causal machine learning and its unique use to emulate randomized controlled trials. Join us to understand how we at Diagnostic Robotics are building models that benefit the patients and help to dramatically reduce the cost of healthcare around the world.