Data-driven, AI-based clinical practice: experiences, challenges, and research directions

Clinical practice is evolving rapidly, away from the traditional but inefficient detect-and-cure approach, and towards a Preventive, Predictive, Personalised and Participative (P4) vision that focuses on extending people’s wellness state. This vision is increasingly data-driven, AI-based, and is underpinned by many forms of "Big Health Data" including periodic clinical assessments and electronic health records, but also using new forms of self-assessment, such as mobile-based questionnaires and personal wearable devices. Over the last few years, we have been conducting a fruitful research collaboration with the Infectious Disease Clinic of the University Hospital of Modena having the main aim of exploring specific opportunities offered by data-driven AI-based approaches to support diagnosis, hospital organization and clinical research. Drawing from this experience, in this paper we provide an overview of the main research challenges that need to be addressed to design and implement data-driven healthcare applications. We present concrete instantiations of these challenges in three real-world use cases and summarise the specific solutions we devised to address them and, finally, we propose a research agenda that outlines the future of research in this field.