Detection of bursting in cultured neuronal networks

In this paper we outline a general framework for detecting cell bursting in recordings of cultured cells or networks. The methodology presents general models for the bursting and the quiescent period. An adaptive Bayesian scheme is detailed which allows the initiation of the bursting to be accurately detected and some prominent features of the burst to be characterized. The proposed framework is mathematically rigorous and uses both the magnitude and phase information of the signal whereas the standard threshold methods commonly used for this purpose are somewhat ad hoc and only utilize the signal magnitude information. Results are presented which illustrate the usefulness of the technique