Spike Detection Using FRI Methods and Protein Calcium Sensors: Performance Analysis and Comparisons

Fast and accurate detection of action potentials from neurophysiological data is key to the study of information processing in the nervous system. Previous work has shown that finite rate of innovation (FRI) theory can be used to successfully reconstruct spike trains from noisy calcium imaging data. This is due to the fact that calcium imaging data can be modeled as streams of decaying exponentials which are a subclass of FRI signals. Recent progress in the development of genetically encoded calcium indicators (GECIs) has produced protein calcium sensors that exceed the sensitivity of the synthetic dyes traditionally used in calcium imaging experiments. In this paper, we compare the suitability for spike detection of the kinetics of a new family of GECIs (the GCaMP6 family) with the synthetic dye Oregon Green BAPTA-1. We demonstrate the high performance of the FRI algorithm on surrogate data for each calcium indicator and we calculate the Cramer-Rao lower bound on the uncertainty of the position of a detected spike in calcium imaging data for each calcium indicator.