New information from old signals: A novel approach to the analysis of physiological waveforms

Huge volumes of physiological waveform data, such as blood pressure and electrocardiogram (ECG) signals, are collected in the diagnosis, treatment and monitoring of patients. Much current analysis focuses only on specific points in these waveforms, thus discarding much useful clinical information. How can we use mathematics to enable us to visualise and quantify changes in the entire waveform?

In this session, we will introduce a novel approach to the technique of delay coordinate embedding of a time series which allows the reconstruction of a two-dimensional attractor using all of the physiological waveform data. Specific development of the method for ECG signals will be discussed, and the potential clinical utility of the resulting attractors will be demonstrated.