Self-organising discovery, recognition and prediction of haemodynamic patterns in the intensive care unit.

To care properly for critically ill patients in the intensive care unit (ICU), clinicians must be aware of haemodynamic patterns. In a typical ICU, a variety of physiological measurements are made continuously and intermittently in an attempt to provide clinicians with the most accurate and precise data needed for recognising such patterns. However, the data are disjointed, yielding little information beyond that provided by instantaneous high/low limit checking. Although instantaneous limit checking is useful for determining immediate dangers, it does not provide much information about temporal patterns. As a result, the clinician is left to sift manually through an excess of data in the interest of generating information. In the study, an arrangement of self-organising artificial neural networks is used to automate the discovery, recognition and prediction of haemodynamic patterns in real time. It is shown that the network is capable of recognising the same haemodynamic patterns recognised by an expert system, DYNASCENE, without being explicitly programmed to do so. Consequently, the system is also capable of discovering new haemodynamic patterns. Results from real clinical data are presented.