This chapter offers a comprehensive exploration of the diverse spectrum of medical data types, including medical imaging, genomics, and electronic health records (EHRs). In doing so, it delves into the anatomy of these data sources and examines their relevance in machine learning applied to clinical decision-making and medical research. Emphasizing the importance of effective data visualization techniques, the chapter highlights how advanced analytics—such as machine learning and artificial intelligence—can be leveraged to uncover new insights, enhance predictive capabilities, and optimize treatment strategies, topics which will be expanded upon in subsequent chapters. By taking a first-principles approach, the chapter investigates how data is derived from the human body, starting at the molecular level and gradually zooming out to the cellular, organ, multi-organ, whole organism, and inter-organism levels. This reductionist framework assumes that a healthy state results from the optimal configuration and location of every molecule. The chapter explores the tools available for interrogating these data at each level, with an emphasis on the time-series nature of the data across these scales, thus constructing a mental framework that can be extended to consider future yet-unknown data sources about the human body. In addition, it explores how effective visualization tools are integral to transforming raw data into actionable knowledge, enabling clinicians to make informed decisions and researchers to draw meaningful conclusions. Finally, the chapter concludes with a discussion on the practicalities and challenges associated with analysing complex medical data, particularly regarding the integration of these diverse data sources and the future of medical data science.