Panel 2 Position Paper: AI could Solve the World's Healthcare Problems and that too at Scale!

Abstract

Artificial Intelligence (AI), and digital technology in general, has the potential to transform healthcare in every setting, and even more in the developing world, where the current healthcare system simply cannot scale to the growing masses, often in more remote locations. Digital technology - mobile phones, remote sensing, and the Internet - can help improve quality of healthcare delivery, reduce cost of healthcare systems, and improve accessibility. However, in addition to AI, there is the need for an appropriate business model for solving healthcare problems. Digital technologies also have the potential to deliver preventative care, focusing on wellness, in addition to personalized disease treatment. In that context, mobile health platforms are more versatile because they can monitor events at a finer granularity and are often used by 'healthy' individuals to track their daily fitness-centric activities. Further, molecular data is available at unprecedented levels and encompass different aspects of the 'being', be in genomic, epigenomic, proteomic, or metagenomic datasets, in other words offering a holistic view of the individual. This can then be integrated with fine-grained physiological signals obtained through wearables. Thus, there are volumes of datasets that need to be annotated, often in multiple formats, offering a multi-dimensional view of the healthy or diseased human. In fact, the human reference genome should have multiple representations based on the race, ethnicity, and the general well-being of the human. This calls for monitoring healthy people at scale, not just diseased people so we get a better understanding of the 'ground truth' for the healthy human being. Overall, AI can therefore be thought of as augmented intelligence that can leverage both better observational and interventional capabilities for delivering precision healthcare and augmented monitoring capabilities to the masses at scale.