Overview of CARMEM: a new dynamic quantitative cardiac model for ECG monitoring and its adaptation to observed signals.

Different approaches have been proposed in order to achieve knowledge integration for coronary care monitoring applications, usually in the form of expert systems. The clinical impact of these expert systems, which are based only on "shallow" knowledge, has not been remarkable due to the difficulties associated with the construction and maintenance of a complete knowledge base. Model-based systems represent an alternative to these problems because they allow efficient integration of the "deep" knowledge on the underlying physiological phenomena being monitored. In this work, a brief review of existing model-based systems for cardiac rhythm interpretation is presented, followed by the description of a new system for Cardiac Arrhythmia Recognition by Model-Based ECG Matching (CARMEM). Fundamental characteristics of CARMEM are presented; in particular, its ability to provide online parameter adaptation to simulate complex rhythms and to match observed ECG signals. The proposed model can be useful for the explanation of the origin of cardiac arrhythmias and contribute towards their robust characterization in the context of coronary care units.