# SCALPEL: Extracting Neurons from Calcium Imaging Data

## TL;DR

This paper introduces SCALPEL, a novel method utilizing dictionary learning and sparse group lasso to automatically identify and analyze neurons in calcium imaging videos, facilitating large-scale neural data analysis.

## Contribution

The paper presents a new dictionary learning-based approach for neuron extraction from calcium imaging data, implemented in an accessible R package.

## Key findings

- Successfully applied to three calcium imaging datasets.
- Provides accurate neuron identification and activity estimation.
- Enables scalable analysis of large neural populations.

## Abstract

In the past few years, new technologies in the field of neuroscience have made it possible to simultaneously image activity in large populations of neurons at cellular resolution in behaving animals. In mid-2016, a huge repository of this so-called "calcium imaging" data was made publicly-available. The availability of this large-scale data resource opens the door to a host of scientific questions, for which new statistical methods must be developed. In this paper, we consider the first step in the analysis of calcium imaging data: namely, identifying the neurons in a calcium imaging video. We propose a dictionary learning approach for this task. First, we perform image segmentation to develop a dictionary containing a huge number of candidate neurons. Next, we refine the dictionary using clustering. Finally, we apply the dictionary in order to select neurons and estimate their corresponding activity over time, using a sparse group lasso optimization problem. We apply our proposal to three calcium imaging data sets. Our proposed approach is implemented in the R package scalpel, which is available on CRAN.