The convergence of AI and machine learning (ML), electronic health records (EHRs), the Internet of Things (IoT), and enhanced data transfer and accessibility has within the last decade delivered a new paradigm in healthcare known as Healthcare 4.0. This rapid transformation to a new digital age in medicine promises new approaches to clinical diagnosis, prediction, decision-making, personalised healthcare and remote patient monitoring. In this chapter, we describe some of the history behind ML in healthcare and review the ML-Big Data nexus, the taxonomy of ML (supervised, unsupervised and reinforcement Learning), the fundamental differences between the fields of statistics and ML, the usual workflow used to develop and validate ML algorithms and how the predictive accuracy of ML algorithms is evaluated. We also discuss the barriers towards the implementation of ML algorithms for clinical decision support within healthcare including ethical considerations, data governance and security, clinician and patient confidence, transparency, data bias, as well as the issues preventing a more rapid integration of AI into healthcare. Finally, we describe some of the more recent developments of ML in healthcare, including quantum ML, federated learning, automated ML, natural language processing and the new progress made in precision medicine via a renewed focus on using reinforcement learning. Throughout the text, we try as far as possible to map the various algorithms and architectures of ML to research questions and healthcare applications with a focus on the older patient population.