Fast deep learning correspondence for neuron tracking and identification in C.elegans using synthetic training

TL;DR

This paper introduces fDLC, a fast deep learning method using transformer networks trained on synthetic data to accurately track and identify neurons in C. elegans across individuals and time, without needing animal straightening.

Contribution

The novel approach employs synthetic training data and transfer learning with transformer networks for neuron tracking and identification in C. elegans.

Findings

Achieves 80% within-animal neuron tracking accuracy

Reaches 65.8% accuracy across animals using position data

Improves to 76.5% accuracy with NeuroPAL color information

Abstract

We present an automated method to track and identify neurons in C. elegans, called "fast Deep Learning Correspondence" or fDLC, based on the transformer network architecture. The model is trained once on empirically derived synthetic data and then predicts neural correspondence across held-out real animals via transfer learning. The same pre-trained model both tracks neurons across time and identifies corresponding neurons across individuals. Performance is evaluated against hand-annotated datasets, including NeuroPAL [1]. Using only position information, the method achieves 80.0% accuracy at tracking neurons within an individual and 65.8% accuracy at identifying neurons across individuals. Accuracy is even higher on a published dataset [2]. Accuracy reaches 76.5% when using color information from NeuroPAL. Unlike previous methods, fDLC does not require straightening or transforming the animal into a canonical coordinate system. The method is fast and predicts correspondence in 10 ms making it suitable for future real-time applications.