Large-scale automated identification of mouse brain cells in confocal light sheet microscopy images

Abstract

Motivation: Recently, confocal light sheet microscopy has enabled high-throughput acquisition of whole mouse brain 3D images at the micron scale resolution. This poses the unprecedented challenge of creating accurate digital maps of the whole set of cells in a brain.Results: We introduce a fast and scalable algorithm for fully automated cell identification. We obtained the whole digital map of Purkinje cells in mouse cerebellum consisting of a set of 3D cell center coordinates. The method is accurate and we estimated an F1 measure of 0.96 using 56 representative volumes, totaling 1.09 GVoxel and containing 4138 manually annotated soma centers.Availability and implementation: Source code and its documentation are available at http://bcfind.dinfo.unifi.it/. The whole pipeline of methods is implemented in Python and makes use of Pylearn2 and modified parts of Scikit-learn. Brain images are available on request.Contact: paolo.frasconi@unifi.itSupplementary information: Supplementary data are available at Bioinformatics online.

Comprehensive optical and data management infrastructure for high-throughput light-sheet microscopy of whole mouse brains

“…28 The prospect of high quality images is especially essential for high throughput imaging methods for whole-brain 3-D reconstructions because the size of datasets is so large that they warrant fully automated postprocessing, for example, for automated cell counting and identification. 7 The effect of optical aberrations can severely obscure features even above the resolution limit and therefore hinder automated segmentation of the data. We have mounted the camera on a separate breadboard, which leaves us with much free, infinity-corrected space in the detection path.…”

“…5,6 Mapping and understanding of this "big data" is an immense task that requires the expertize of computer scientists to employ fully automated, scalable postprocessing, for example, to carry out blood vessel segmentation or cell counting. 7 On the other hand, the imaging of large, intrinsically opaque samples in LSFM necessitates clearing protocols based on refractive index matching, which render the tissue transparent. 8,9 Therefore, technological advances in LSFM need to be matched by novel development in the area of information and biotechnology.…”

BioImage Informatics: The challenge of knowledge extraction from biological images

“…However, while computer vision typically tries to interpret images or videos of complex scenes, possibly with people and with very little a-priori assumption about the objects to be analyzed [13], bioimage informatics work with images with a low signal-tonoise ratio and limited resolution. Fortunately, these difficulties may be unravelled to a certain extent by a larger simplicity of the scene and by the injection of a-priori biological information [14].…”

“…Among the numerous applications of bioimage informatics, recent efforts in this field have been directed towards the neuroscience with the ambitious goal of understanding how human brain works using anatomical, molecular and functional maps [3], [4], [5], [6], [14], [15]. In particular, an impelling need is the comprehension of the cytoarchitecture of the central nervous system on a brain-wide scale at the different scales [16], [17].To image these structures acquisition devices with micro-resolution are needed.…”

“…Although a recent work still manually identifies dendritic spine in rat cerebellum acquired at the laser confocal microscope, thus being prone to oversight human error and limited to not so large 3D volume [18], there are recent efforts to deploy automatic tools for automatic image analysis that are, however, limited to relatively small volumes [19], [20], [21], [22], [23]. Moreover, in the literature few works deal with terabytes-sized datasets [3], [14], [24]: they reveal that not only tools for large scale image processing are required, but there is a compelling need of a new neuroinformatics framework for low and high level data analysis, allowing to detect statistical regularities and the relationships between data at different levels of brain organization.…”