Calcium imaging is an important technique for monitoring the activity of thousands of neurons simultaneously. As calcium imaging datasets grow in size, automated detection of individual neurons is becoming important. Here we apply a supervised learning approach to this problem and show that convolutional networks can achieve near-human accuracy and superhuman speed. Accuracy is superior to the popular PCA/ICA method based on precision and recall relative to ground truth annotation by a human expert. These results suggest that convolutional networks are an efficient and flexible tool for the analysis of large-scale calcium imaging data.
@article{arxiv.1606.07372,
title = {Automatic Neuron Detection in Calcium Imaging Data Using Convolutional Networks},
author = {Noah J. Apthorpe and Alexander J. Riordan and Rob E. Aguilar and Jan Homann and Yi Gu and David W. Tank and H. Sebastian Seung},
journal= {arXiv preprint arXiv:1606.07372},
year = {2016}
}
Comments
9 pages, 5 figures, 2 ancillary files; minor changes for camera-ready version. appears in Advances in Neural Information Processing Systems 29 (NIPS 2016)