Touching Soma Segmentation Based on the Rayburst Sampling Algorithm

Abstract

Neuronal soma segmentation is essential for morphology quantification analysis. Rapid advances in light microscope imaging techniques have generated such massive amounts of data that time-consuming manual methods cannot meet requirements for high throughput. However, touching soma segmentation is still a challenge for automatic segmentation methods. In this paper, we propose a soma segmentation method that combines the Rayburst sampling algorithm and ellipsoid fitting. The improved Rayburst sampling algorithm is used to detect the soma surface; the ellipsoid fitting method then refines jagged sampled soma surface to generate smooth ellipsoidal shapes for efficient analysis. In experiments, we validated the proposed method by applying it to datasets from the fluorescence micro-optical sectioning tomography (fMOST) system. The results indicate that the proposed method is comparable to the manual segmented gold standard with accurate soma segmentation at a relatively high speed. The proposed method can be extended to large-scale image stacks in the future.Electronic supplementary materialThe online version of this article (10.1007/s12021-017-9336-y) contains supplementary material, which is available to authorized users.

Cell numbers, distribution, shape, and regional variation throughout the murine hippocampal formation from the adult brain Allen Reference Atlas

“…Leaping progress in image recognition and the continuous growth in computing power have substantially altered this status quo (Bhanu and Peng 2000;Peng et al 2013). Specifically, several algorithms were recently designed to enable high-throughput soma detection and analysis (Hu et al 2017;Kayasandik and Labate 2016;Luengo-Sanchez et al 2015;Tapias and Greenamyre 2014;Zhang et al 2018;Quan et al 2013). Fully automatic cell segmentation modules are also implemented in freely available mainstream software programs such as ImageJ (Schindelin et al 2015;Schneider et al 2012) and CellProfiler (Bray et al 2015;Lamprecht et al 2007), allowing robust quantification of soma count, location, and geometry in large-scale applications.…”

Accurate Neuronal Soma Segmentation Using 3D Multi-Task Learning U-Shaped Fully Convolutional Neural Networks

“…In this study, we compare the proposed method with several 3D neuronal soma segmentation methods in Nissl-stained dataset, including the concave points clustering (CPC) random walker algorithm (He et al, 2014 ), the distance transform-based rayburst sampling algorithm (Hu et al, 2017 ), and 3D FCNs (3D UNet, Vnet) (Çiçek et al, 2016 ; Milletari et al, 2016 ). To validate the deep learning results similarly, all the methods are evaluated using a 3-fold cross-validation, where the dataset is split into three groups (seven images, seven images, six images).…”

“…The rayburst sampling algorithm displayed over-segmentation and under-segmentation in the Nissl-stained dataset. It is thought that the ellipsoid model could not accurately describe the irregular-shaped neuronal somata, and may have split the elongated somata into multiple ones or missed the plat-shaped somata (Hu et al, 2017 ). These methods assume that the neuronal soma is ball-like or ellipsoid-like, which may not suit irregular-shaped neuronal somata.…”

Soma Segmentation Using U-Shaped Convolutional Neural Networks with Weight Boundary Mechanism

“…The study of neuronal morphology is a very important branch in the whole field of brain science. Neuronal morphology facilitates the understanding of the brain as a complex network [1], learning the connections between brain structure and function [2], classification of neurons [3], dynamic analysis [4], etc. Soma morphology can be used to analyze neuronal morphology qualitatively and quantitatively [2].…”

“…Neuronal morphology facilitates the understanding of the brain as a complex network [1], learning the connections between brain structure and function [2], classification of neurons [3], dynamic analysis [4], etc. Soma morphology can be used to analyze neuronal morphology qualitatively and quantitatively [2]. And the soma segmentation out is more conducive to neuronal reconstruction, which makes the reconstructed neuronal structure more accurate [5] and more conducive to the study of neuronal morphology.…”