Towards integration of artificial intelligence into medical devices as a real-time recommender system for personalised healthcare: State-of-the-art and future prospects

Abstract

In the era of big data, artificial intelligence (AI) algorithms have the potential to revolutionize healthcare by improving patient outcomes and reducing healthcare costs. AI algorithms have frequently been used in health care for predictive modelling, image analysis and drug discovery. Moreover, as a recommender system, these algorithms have shown promising impacts on personalized healthcare provision. A recommender system learns the behaviour of the user and predicts their current preferences (recommends) based on their previous preferences. Implementing AI as a recommender system improves this prediction accuracy and solves cold start and data sparsity problems. However, most of the methods and algorithms are tested in a simulated setting which cannot recapitulate the influencing factors of the real world. This review article systematically reviews prevailing methodologies in recommender systems and discusses the AI algorithms as recommender systems specifically in the field of healthcare. It also provides discussion around the most cutting-edge academic and practical contributions present in the literature, identifies performance evaluation matrices, challenges in the implementation of AI as a recommender system, and acceptance of AI-based recommender systems by clinicians. The findings of this article direct researchers and professionals to comprehend currently developed recommender systems and the future of medical devices integrated with real-time recommender systems for personalized healthcare.