Deep learning method for comet segmentation and comet assay image analysis

Yiyu Hong,Hannah Lee,Ji-Young Lee,Woo-Chan Son,Hee Seon Lim,Insuk Sohn,Donghwan Lee,Ju-Hyung Seok,Hyo-Jeong Han,Zhen-Yu Hong,Junsu Ko

doi:10.1038/s41598-020-75592-7

Yiyu Hong, Hannah Lee + Show 9 more

Open Access

https://doi.org/10.1038/s41598-020-75592-7

Copy DOI

Abstract

Comet assay is a widely used method, especially in the field of genotoxicity, to quantify and measure DNA damage visually at the level of individual cells with high sensitivity and efficiency. Generally, computer programs are used to analyze comet assay output images following two main steps. First, each comet region must be located and segmented, and next, it is scored using common metrics (e.g., tail length and tail moment). Currently, most studies on comet assay image analysis have adopted hand-crafted features rather than the recent and effective deep learning (DL) methods. In this paper, however, we propose a DL-based baseline method, called DeepComet, for comet segmentation. Furthermore, we created a trainable and testable comet assay image dataset that contains 1037 comet assay images with 8271 manually annotated comet objects. From the comet segmentation test results with the proposed dataset, the DeepComet achieves high average precision (AP), which is an essential metric in image segmentation and detection tasks. A comparative analysis was performed between the DeepComet and the state-of-the-arts automatic comet segmentation programs on the dataset. Besides, we found that the DeepComet records high correlations with a commercial comet analysis tool, which suggests that the DeepComet is suitable for practical application.

Highlights

Several image analysis methods, such as Comet Score, CASP, OpenComet, CometQ, and HiComet for comet assay have been proposed to facilitate the comet scoring p rocess[16,17,18,19,20]
We proposed DeepComet for comet segmentation that utilizes the Mask R-convolutional neural network (CNN) architecture
Developed by Facebook AI research, the Mask R-CNN is one of the most popular architectures because it can be extended to other applications such as human pose estimation

Summary

Introduction

Several image analysis methods, such as Comet Score, CASP, OpenComet, CometQ, and HiComet for comet assay have been proposed to facilitate the comet scoring p rocess[16,17,18,19,20]. The primary performance differences come from whether a tool can detect and segment each comet correctly rather than scoring stage because the metric used to assess the comet score has predefined basic standards and r ules[21]. These image analysis methods all use hand-crafted image features with traditional machine learning (such as support vector machine) at comet detection and segmentation stage. A new fully-automated comet assay analysis program (DeepComet) using a DL method (rather than hand-crafted features) is proposed to improve the effectiveness of comet image analysis. The performance of the proposed DL-based baseline method (DeepComet) compared to those of state-of-the-art comet segmentation programs such as OpenComet, HiComet, Comet Assay IV is discussed

Methods

Results

Conclusion