Section: New Results

Classification of Cardiac Arrhythmia in vitro based on Multivariate Complexity Analysis

Participant : Binbin Xu.

Physiological signals are temporal series containing a lot of information, and their analysis (either for diagnosis or evolution monitoring) necessitates tools that take into account their intrinsic characteristics, notably in terms of impredictability and high number of parameters. Methodologies coming from chaotic and nonlinear dynamical systems contain some useful building blocks in that perspective, and allow a qualitative link with phenomenological and bio-inspired models. The objective of this work is to introduce some methods in nonlinear dynamics useful for the processing of these types of signals. An application of these tools is illustrated in the processing of potential electrical fields acquired from in vitro culture cells on newborn rats. Both normal (regular contraction of cells) and arrythmic (disordoned contractions) cases are contemplated.

Figure 2. CPE period bifurcartion diagram. Cells are stimulated by M1 electrode during 5 minutes with an impulsion train of $200\mu m$ and frequency $100\mathrm{Hz}$. Three particular phenomena in cell behaviour: A ($t=1$ hour) chaotic state, B ($t=3$ hour) and C ($t=3.5$ hour) period doubling phase, D ($t=4$ hour) regular and stable rythm.

The bifurcation diagram is an example of a tool that can be used in the temporal analysis of an experimental system.

Publication: [32] .