A Deep Learning-Based Segmentation of Cells and Analysis (DL-SCAN)

Abstract

With the recent surge in the development of highly selective probes, fluorescence microscopy has become one of the most widely used approaches to study cellular properties and signaling in living cells and tissues. Traditionally, microscopy image analysis heavily relies on manufacturer-supplied software, which often demands extensive training and lacks automation capabilities for handling diverse datasets. A critical challenge arises, if fluorophores employed exhibit low brightness and low Signal-to-Noise ratio (SNR). As a consequence, manual intervention may become a necessity, introducing variability in the analysis outcomes even for identical samples when analyzed by different users. This leads to the incorporation of blinded analysis which ensures that the outcome is free from user bias to a certain extent but is extremely time-consuming. To overcome these issues, we have developed a tool called DL-SCAN that automatically segments and analyzes fluorophore-stained regions of interest such as cell bodies in fluorescence microscopy images using a Deep Learning algorithm called Stardist. We demonstrate the programs ability to automate cell identification and study cellular ion dynamics using synthetic image stacks with varying SNR. This is followed by its application to experimental Na+ and Ca2+ imaging data from neurons and astrocytes in mouse brain tissue slices exposed to transient chemical ischemia. The results from DL-SCAN are consistent, reproducible, and free from user bias, allowing efficient and rapid analysis of experimental data in an objective manner. The open-source nature of the tool also provides room for modification and extension to analyze other forms of microscopy images specific to the dynamics of different ions in other cell types. Statement of SignificanceFluorescence microscopy is widely used to study the functional and morphological features of living cells. However, various factors, such as low SNR, background noise, drift in the signal, movement of the tissue, and the large size of the resulting imaging data, make the processing of fluorescence microscopy data prone to errors, user bias, and extremely time-consuming. These and other issues hinder the full utilization of these powerful experimental techniques. Our novel Deep Learning-based tool overcomes these issues by processing and analyzing fluorescence imaging data, e.g., enabling automated visualization of ion changes in living cells in brain slices. Yet the tool remains easy to use with a streamlined workflow.