Neural networks and expert systems in acute & primary medicine - reflections on a decade of research studies

The paper describes studies undertaken in the use of artificial intelligence (AI) technologies in decision-making in (a) the selection of pacemakers, and in the clinical management of (a) asthma and (b) osteoporosis. The specific studies described include an on-going study in a health authority in the UK and earlier studies undertaken in the University of London at St George's Hospital Medical School within acute medicine and primary care medicine. The first investigations were carried out at St George's Hospital Medical School and were in acute medicine e.g. investigations of the potential of expert systems for post infarct risk stratification and for the selection of arificial pacemakers [10, 11]. The investigations then moved on to primary care medicine exploring the potential of neural networks in asthma screening [8, 9]. More recently, a study has been made of neural networks in the clinical management of osteoporosis. This paper reflects on the lessons gained from these AI activities which have lasted over a decade.