FAST: Fast, free, consistent, and unsupervised oligodendrocyte segmentation and tracking system.

To develop reparative therapies for neurological disorders like multiple sclerosis (MS), we need to better understand the physiology of loss and replacement of oligodendrocytes, the cells that make myelin and are the target of damage in MS. In vivo two-photon fluorescence microscopy allows direct visualization of oligodendrocytes in the intact brain of transgenic mouse models, promising a deeper understanding of the longitudinal dynamics of replacing oligodendrocytes after damage. However, the task of tracking the fate of individual oligodendrocytes requires extensive effort for manual annotation and is especially challenging in three-dimensional images. While several models exist for annotating cells in two-dimensional images, few models exist to annotate cells in three-dimensional images and even fewer are designed for tracking cells in longitudinal imaging. Notably, existing options often come with a substantial financial investment, being predominantly commercial or confined to proprietary software. Furthermore, the complexity of processes and myelin formed by individual oligodendrocytes can result in the failure of algorithms that are specifically designed for tracking cell bodies alone. Here, we propose a fast, free, consistent, and unsupervised beta-mixture oligodendrocyte segmentation system (FAST) that is written in open-source software, and can segment and track oligodendrocytes in three-dimensional images over time with minimal human input. We showed that the FAST model can segment and track oligodendrocytes similarly to a blinded human observer. Although FAST was developed to apply to our studies on oligodendrocytes, we anticipate that it can be modified to study four-dimensional in vivo data of any brain cell with associated complex processes. We have developed "FAST: Fast, free, consistent, and unsupervised oligodendrocyte segmentation and tracking system" to solve our challenge of quantification of four-dimensional data acquired from longitudinal in vivo imaging. Although it was developed for oligodendrocytes, we will make the code entirely open source and user-friendly, and expect that it will be useful for segmentation for any cell body from a complex cell amenable to longitudinal in vivo imaging.