Different cell imaging methods did not significantly improve immune cell image classification performance.

Taisaku Ogawa,Tomokatsu Ikawa,Tomoharu Iwata,Koji Ochiai,Taku Tsuzuki,Katsuyuki Shiroguchi,Koichi Takahashi,Chi-Hua Chen

doi:10.1371/journal.pone.0262397

Taisaku Ogawa, Tomokatsu Ikawa + Show 6 more

Open Access

https://doi.org/10.1371/journal.pone.0262397

Copy DOI

Abstract

Developments in high-throughput microscopy have made it possible to collect huge amounts of cell image data that are difficult to analyse manually. Machine learning (e.g., deep learning) is often employed to automate the extraction of information from these data, such as cell counting, cell type classification and image segmentation. However, the effects of different imaging methods on the accuracy of image processing have not been examined systematically. We studied the effects of different imaging methods on the performance of machine learning-based cell type classifiers. We observed lymphoid-primed multipotential progenitor (LMPP) and pro-B cells using three imaging methods: differential interference contrast (DIC), phase contrast (Ph) and bright-field (BF). We examined the classification performance of convolutional neural networks (CNNs) with each of them and their combinations. CNNs achieved an area under the receiver operating characteristic (ROC) curve (AUC) of ~0.9, which was significantly better than when the classifier used only cell size or cell contour shape as input. However, no significant differences were found between imaging methods and focal positions.

Highlights

Recent advances in automated microscopy have made it possible to collect large numbers of cell images [1]
We evaluated the dependence of classification performance on the imaging methods using the area under the curve (AUC) of the receiver operating characteristic (ROC) as a performance measure
We checked the classification performance of convolutional neural networks (CNNs) by the following five investigations: 1) comparison of the AUC between the cases using only cell size or cell shape and where cell images were given as input; 2) the AUC when multi-channel inputs with imaging methods were given; 3) variation of the AUC when shifting the focus; 4) variation of the AUC when changing the amount of training data; and 5) variation of the AUC depending on the initial random number

Summary

Introduction

Recent advances in automated microscopy have made it possible to collect large numbers of cell images [1]. It is becoming increasingly difficult to analyse these large amounts of data manually. The automation of judgements by machine learning is expected to improve the speed and processing of large amounts of data and ensure consistency in the results of judgements [2, 3]. The application of machine learning to biological image processing is expanding with the development of deep learning [4,5,6]. An interesting use of deep learning in cell biological research is to infer the differentiation of living cells.

Methods

Results

Conclusion