Center-Extraction-Based Three Dimensional Nuclei Instance Segmentation of Fluorescence Microscopy Images

TL;DR

This paper introduces a novel two-stage CNN-based method for 3D nuclei segmentation in fluorescence microscopy images, utilizing synthetic data generation to overcome labeling challenges and achieving superior performance on real datasets.

Contribution

It presents a new approach combining synthetic data generation with a center-extraction-based segmentation method for 3D nuclei in microscopy images.

Findings

Outperforms existing segmentation techniques on real microscopy datasets

Uses synthetic volumes generated by a cycle-consistent adversarial network

Effective for large and complex 3D microscopy data

Abstract

Fluorescence microscopy is an essential tool for the analysis of 3D subcellular structures in tissue. An important step in the characterization of tissue involves nuclei segmentation. In this paper, a two-stage method for segmentation of nuclei using convolutional neural networks (CNNs) is described. In particular, since creating labeled volumes manually for training purposes is not practical due to the size and complexity of the 3D data sets, the paper describes a method for generating synthetic microscopy volumes based on a spatially constrained cycle-consistent adversarial network. The proposed method is tested on multiple real microscopy data sets and outperforms other commonly used segmentation techniques.