A deep learning algorithm for 3D cell detection in whole mouse brain image datasets

Abstract

Understanding the function of the nervous system necessitates mapping the spatial distributions of its constituent cells defined by function, anatomy or gene expression. Recently, developments in tissue preparation and microscopy allow cellular populations to be imaged throughout the entire rodent brain. However, mapping these neurons manually is prone to bias and is often impractically time consuming. Here we present an open-source algorithm for fully automated 3D detection of neuronal somata in mouse whole-brain microscopy images using standard desktop computer hardware. We demonstrate the applicability and power of our approach by mapping the brain-wide locations of large populations of cells labeled with cytoplasmic fluorescent proteins expressed via retrograde trans-synaptic viral infection.

The mouse brain after foot shock in four dimensions: Temporal dynamics at a single-cell resolution

“…Of note, we excluded from the analysis six brain areas because we deemed their size too small for a reliable quantification ( SI Appendix , Table S1 ). Until this step, we adapted tools developed by others [Imaris, Elastix ( 23 ), and Clearmap ( 8 )]; alternative tools [e.g., CellFinder ( 24 ) and WholeBrain ( 25 )] could have also been used for the same purpose.…”

“…This type of design is essential for effective batch effect correction. The data cleaning (step 2) and preprocessing pipelines (step 3) are available for similar future questions in the R package developed for the purpose, abc4d (“Analysis Brain Cellular activation in 4 Dimensions”), which is interoperable with several annotation/alignment tools [step 1; e.g., Clearmap ( 8 ) and CellFinder ( 24 )].…”

Accurate determination of marker location within whole-brain microscopy images

“…Since they are built upon open-source Python data analysis and visualisation tools including the BrainGlobe Atlas API 4 they are compatible with multiple existing, as well as future brain atlases. Because the tools in the BrainGlobe suite are compatible with these atlases, the results of brainreg-segment can be directly compared with results from other software, for example cellfinder, to compare the distribution of individually labelled cells to a bulk injection site in physiological conditions 15 or after experimentally-induced trauma 16 . Also, these tools were developed as plugins for napari to provide interoperability with other analysis software, and importantly to streamline installation and use.…”

Tools for efficient analysis of neurons in a 3D reference atlas of whole mouse spinal cord

“…Atlases provide a high-resolution framework for comparative analyses and have enabled massive collaborative projects like the BRAIN initiative ( Ecker et al, 2017 ). Various computational tools have been developed for the analysis and integration of individual brain datasets ( Bakker et al, 2015 ; Botta et al, 2020 ; Chon et al, 2019 ; Eastwood et al, 2019 ; Friedmann et al, 2020 ; Oh et al, 2014 ; Puchades et al, 2019 ; Shiffman et al, 2018 ; Tappan et al, 2019 ; Tyson et al, 2021 ; Wang et al, 2021 ), improving both speed and quality of experiments. Together, these resources have enabled brain-wide mapping studies of cell types and neuronal connectivity and dramatically accelerated scientific discovery in this field of research.…”

“…A second problem is that, regardless of acquisition method, resources for analyzing whole-SC data are scarce and lag far behind the manifold tools available for whole-brain reconstruction, atlas registration, and data interpretation ( Bakker et al, 2015 ; Botta et al, 2020 ; Chon et al, 2019 ; Eastwood et al, 2019 ; Friedmann et al, 2020 ; Oh et al, 2014 ; Puchades et al, 2019 ; Shiffman et al, 2018 ; Tappan et al, 2019 ; Tyson et al, 2021 ; Wang et al, 2021 ) . 3D reconstruction and analysis of human SC MRI data has been reported, but the tools offer only low-resolution data registration of larger gray matter (GM) and white matter (WM) regions ( De Leener et al, 2017 ; Prados et al, 2016 ).…”