Internet-of-Things and machine learning promise a new era for healthcare. The emergence of transformative technologies, such as Implantable and Wearable Medical Devices (IWMDs), has enabled collection and analysis of physiological signals from anyone anywhere anytime. Machine learning allows us to unearth patterns in these signals and make healthcare predictions in both daily and clinical situations. This broadens the reach of healthcare from conventional clinical contexts to pervasive everyday scenarios, from passive data collection to active decision-making. Despite the existence of a rich literature on IWMD-based and clinical healthcare systems, the fundamental challenges associated with design and implementation of smart healthcare systems have not been well-addressed. The main objectives of this article are to define a standard framework for smart healthcare aimed at both daily and clinical settings, investigate state-of-the-art smart healthcare systems and their constituent components, discuss various considerations and challenges that should be taken into account while designing smart healthcare systems, explain how existing studies have tackled these design challenges, and finally suggest some avenues for future research based on a set of open issues and challenges.