[
International Worm Meeting,
2021]
Analysis of calcium imaging recordings is tedious, repetitive and time consuming. In this new project, we propose to develop a machine learning-based automated tool for the analysis of recordings from C. elegans neurons, thereby reducing noise, time loss and experimenter bias. We are specifically training our model to segment time-lapse fluorescence recordings of the RIA interneuron in semi-restrained animals. Our tool can segment animals and track their head movements as well as identify the three compartments of the RIA neurite. This prototype tool demonstrates the potential of deep learning to accelerate and improve data acquisition from time-lapse fluorescence recordings and other imaging data.