DeepUnitMatch: tracking neurons across days in electrophysiology using Deep Neural Networks

Abstract To understand neural processes such as learning or memory, we need to track the activity of populations of neurons at the level of single spikes and across days. Here, we leverage deep neural networks to build DeepUnitMatch, a software that reliably tracks individual neurons in high-density electrophysiological recordings across weeks. DeepUnitMatch uses only the spike waveforms of the neurons, and not their spiking patterns, and outperforms current solutions. Competing Interest Statement The authors have declared no competing interest. Funder Information Declared Wellcome Trust, https://ror.org/029chgv08, 223144/Z/21/Z, 227065 European Union's Horizon 2020, 101022757 Copyright The copyright holder for this preprint is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license.