# Deep learning detection of dynamic exocytosis events in fluorescence TIRF microscopy

## TL;DR

A deep learning model called ExoDeepFinder was developed to detect dynamic exocytosis events in live-cell microscopy, outperforming traditional methods and adapting well to new conditions.

## Contribution

Adapted a 3D cryo-ET deep learning model for dynamic exocytosis detection in 2D time-lapse fluorescence microscopy.

## Key findings

- ExoDeepFinder outperformed conventional unsupervised methods in detecting exocytosis events.
- The model demonstrated robustness to new experimental conditions without requiring retraining.
- ExoDeepFinder and its training datasets are openly available for further use and adaptation.

## Abstract

Segmentation and detection of biological objects in fluorescence microscopy is of paramount importance in cell imaging. Deep learning approaches have recently shown promise to advance, automatize and accelerate analysis. However, most of the interest has been given to the segmentation of static objects of 2D/3D images whereas the segmentation of dynamic processes obtained from time-lapse acquisitions has been less explored. Here we adapted DeepFinder, a U-Net originally designed for 3D noisy cryo-electron tomography (cryo-ET) data, for the detection of rare dynamic exocytosis events (termed ExoDeepFinder) observed in temporal series of 2D Total Internal Reflection Fluorescence Microscopy (TIRFM) images. ExoDeepFinder achieved good absolute performances with a relatively small training dataset of 12000 events in 60 cells. We rigorously compared deep learning performances with unsupervised conventional methods from the literature. ExoDeepFinder outcompeted the tested methods, but also exhibited a greater plasticity to the experimental conditions when tested under drug treatments and after changes in cell line or imaged reporter. This robustness to unseen experimental conditions did not require re-training demonstrating generalization capability of our deep learning model. ExoDeepFinder, as well as the annotated training datasets, were made transparent and available through an open-source software as well as a Napari plugin and can directly be applied to custom user data. The apparent plasticity and performances of ExoDeepFinder to detect dynamic events open new opportunities for future deep learning guided analysis of dynamic processes in live-cell imaging.

Until now, deep learning in microscopy has been largely restricted to the detection of static biological objects. Here, we adapted a neural network designed for macromolecule detection in 3D electron microscopy to instead detect dynamic exocytosis events from time-lapse fluorescence microscopy, calling it “ExoDeepFinder”. Exocytosis detection is of prime interest because of its involvement in many fundamental physiological processes, from digestive enzyme secretion to neurotransmitter release. Comparing ExoDeepFinder to traditional exocytosis detection methods on a large dataset, we demonstrated the superiority of our deep learning approach. ExoDeepFinder is more robust than other methods, even when faced with experimental conditions unseen during training. Our dataset and method are transparent and freely available, thus enabling re-training and extension for other dynamic cell processes.