Stephen E. Rees is Professor of respiratory and critical care technology at Aalborg University, Denmark. His interests are in the mathematical modelling of physiological processes and their application to solve clinical problems, typically as bed-side decision support tools. He is Editor-in Chief of the Journal of Clinical Monitoring and Computing, a journal dedicated to publishing papers describing technical solutions to clinical problems in anaesthesia, intensive care and peri-operative monitoring.
In recent years the term Artificial Intelligence (AI) has become synonymous with machine learning applied to big data. However, a number of useful clinical tools are often based on rules, or on physiological models combined with decision theory. This talk presents one such system for advising on the correct settings of mechanical ventilation. The system, based on physiological models and decision theory, has transitioned from a laboratory bench system (INVENT) to an FDA approved commercial system (the Beacon Caresystem) currently in large scale randomised control trials. The talk will cover the science included in the system, its evaluation, and the technological and human challenges facing integration of such systems. Questions will be asked as to the nature of AI today, and whether all our knowledge can, is, or should be, encompassed in that which we learn from big data, or whether a combination of methods can act in synergy to improve care.