Fast, Simple Calcium Imaging Segmentation with Fully Convolutional Networks

We present a modular approach for analyzing calcium imaging recordings of large neuronal ensembles. Our goal is to simultaneously identify the locations of the neurons, demix spatially overlapping components, and denoise and deconvolve the spiking activity from the slow dynamics of the calcium indicator. Our approach relies on a constrained nonnegative matrix factorization that expresses the spatiotemporal fluorescence activity as the product of a spatial matrix that encodes the spatial footprint of each neuron in the optical field and a temporal matrix that characterizes the calcium concentration of each neuron over time. This framework is combined with a novel constrained deconvolution approach that extracts estimates of neural activity from fluorescence traces, to create a spatiotemporal processing algorithm that requires minimal parameter tuning. We demonstrate the general applicability of our method by applying it to in vitro and in vivo multi-neuronal imaging data, whole-brain light-sheet imaging data, and dendritic imaging data.

Recent advances in fluorescence imaging permit studies of Ca(2+) dynamics in large numbers of cells, in anesthetized and awake behaving animals. However, unlike for electrophysiological signals, standardized algorithms for assigning optically recorded signals to individual cells have not yet emerged. Here, we describe an automated sorting procedure that combines independent component analysis and image segmentation for extracting cells' locations and their dynamics with minimal human supervision. In validation studies using simulated data, automated sorting significantly improved estimation of cellular signals compared to conventional analysis based on image regions of interest. We used automated procedures to analyze data recorded by two-photon Ca(2+) imaging in the cerebellar vermis of awake behaving mice. Our analysis yielded simultaneous Ca(2+) activity traces for up to >100 Purkinje cells and Bergmann glia from single recordings. Using this approach, we found microzones of Purkinje cells that were stable across behavioral states and in which synchronous Ca(2+) spiking rose significantly during locomotion.