Sequential Bayesian Detection of Spike Activities From Fluorescence Observations

Extracting and detecting spike activities from the fluorescence observations is an important step in understanding how neuron systems work. The main challenge lies in the combined ambient noise with fluctuated baseline, which contaminates the observations, thereby deteriorating the reliability of spike detection. This may be even worse in the face of the nonlinear biological process, the coupling interactions between spikes and baseline, and the unknown critical parameters of an underlying model, in which erroneous estimations of parameters will affect the detection of spikes causing further error propagation. The state-of-the-art MLSpike is premised on static parameter inference on spike events and ignores sequential spike nonlinear interactions. In this paper, we propose a random finite set (RFS) based Bayesian inference approach, which encapsulates the dynamics of sequential spikes, fluctuated baseline, and unknown model parameters. Specifically, the cardinal probability of RFS is able to distinguish latent spike behaviours (e.g., spike or non-spike). Our results demonstrate that the proposed scheme can gain an extra 12% detection accuracy in comparison with the state-of-the-art MLSpike method.