DeepSynth: Three-dimensional nuclear segmentation of biological images using neural networks trained with synthetic data

Kenneth W Dunn,Paul Salama,Soonam Lee,David Joon Ho,Chichen Fu,Edward J Delp,Shuo Han

doi:10.1038/s41598-019-54244-5

Abstract

The scale of biological microscopy has increased dramatically over the past ten years, with the development of new modalities supporting collection of high-resolution fluorescence image volumes spanning hundreds of microns if not millimeters. The size and complexity of these volumes is such that quantitative analysis requires automated methods of image processing to identify and characterize individual cells. For many workflows, this process starts with segmentation of nuclei that, due to their ubiquity, ease-of-labeling and relatively simple structure, make them appealing targets for automated detection of individual cells. However, in the context of large, three-dimensional image volumes, nuclei present many challenges to automated segmentation, such that conventional approaches are seldom effective and/or robust. Techniques based upon deep-learning have shown great promise, but enthusiasm for applying these techniques is tempered by the need to generate training data, an arduous task, particularly in three dimensions. Here we present results of a new technique of nuclear segmentation using neural networks trained on synthetic data. Comparisons with results obtained using commonly-used image processing packages demonstrate that DeepSynth provides the superior results associated with deep-learning techniques without the need for manual annotation.

Highlights

The scale of biological microscopy has increased dramatically over the past ten years, with the development of new modalities supporting collection of high-resolution fluorescence image volumes spanning hundreds of microns if not millimeters
The user provides a set of training data that includes objects that have been outlined by the user, from which a convolutional neural network derives a set of common features that are used to discriminate objects in the experimental data
The DeepSynth segmentation technique was first tested in a 3D fluorescence image volume collected from rat kidney tissue

Summary

Introduction

The scale of biological microscopy has increased dramatically over the past ten years, with the development of new modalities supporting collection of high-resolution fluorescence image volumes spanning hundreds of microns if not millimeters. The size and complexity of these volumes is such that quantitative analysis requires automated methods of image processing to identify and characterize individual cells. For many workflows, this process starts with segmentation of nuclei that, due to their ubiquity, ease-of-labeling and relatively simple structure, make them appealing targets for automated detection of individual cells. For studies involving relatively few measurements, this can be accomplished by manually outlining regions-of-interest, whose fluorescence can be measured using available software This approach quickly becomes impractical as the number of measurements increases and/ or the regions must be defined in three dimensions. The user provides a set of training data that includes objects that have been outlined by the user, from which a convolutional neural network derives a set of common features that are used to discriminate objects in the experimental data

Methods

Results

Conclusion