Determination of the information gain by estimating the instantaneous phase of biosignals

  • Motivation
    In a modern clinical environment, the monitoring of the patients is a common technique to assist the physicians with diagnosis, therapy and risk stratification. During a normal day, thousands of minutes of patient data are recorded. Every day, these need to be evaluated automatically to support the health personell. The appropriate visualisation of the health status of the patient is crucial to let the physician interpret the results quickly. In particular, during the evaluation of temporal 2D or 3D activation sequences, the visualisation of the instantaneous phase of a time series of biosignals is supposed to deliver the physician an additional benefit by enabling him to follow an activation wave or pulse propagation on a picture or volume and can therefore easily assess a localised failure of propagation. Especially in intracardiac electrograms from patients with atrial fibrillation, the phase determination is claimed to improve the diagnosis and therapy since it is a robust detector for activity. Moreover, in other modalities, too, the phase could deliver an additional benefit.


    In this project, the hypothesis “the calculation of the phase delivers an additional benefit and a robust detection method for activity” will be evaluated. In the first part of this research project, a signal processing study using synthetic signals will be carried out with the aim of studying how different phase detection methods perform. This evaluation should be performed on signals mimicking atrial electrograms and signals from the electrical impedance tomography (EIT) including respiration and cardiac activity. With the knowledge gained from the study, also a general purpose method to estimate the instantaneous frequency will be developed and validated.

    In the second part of the project, the clinical relevance of the instantaneous phase will be studied. For this purpose, the phase will be estimated for stable rhythms from atrial signals showing different complexities of signal morphologies. Subsequently, the value of the instantaneous phase should also be assessed for the determination of cardiac activity in signals from EIT.