Decision Support Systems for Healthcare based on Probabilistic Graphical Models: A Survey and Perspective

Abstract

The main purpose of a healthcare support system is to provide timely and accurate information to clinicians, patients, and others to inform decisions about healthcare. Healthcare support systems can potentially lower costs, improve efficiency, and reduce patient inconvenience. For example, Healthcare support systems can help by alerting clinicians about the possible duplicate tests a patient may be about to receive. Owing to the usability of healthcare, decision support systems have been an utmost research focus of academia and industries. Hence, in this chapter, we provide a survey of different decision support systems being used in healthcare. More specifically, we provide a survey about decision support systems based on probabilistic graphical models and machine learning, followed by a future perspective, implications, and challenges. Furthermore, to give an in-depth intuition about probabilistic models, we provide case studies of decision support systems using machine learning and Probabilistic Graphical Models (PGMs) in electrocardiograph (ECG) classification and also show how explainable AI can further improve such systems.