Live-dead assay on unlabeled cells using phase imaging with computational specificity

Chenfei Hu,Edward M Kong,Yuchen He,Shenghua He,Hua Li,Young Jae Lee,Gabriel Popescu,Mark A Anastasio

doi:10.1038/s41467-022-28214-x

Abstract

Existing approaches to evaluate cell viability involve cell staining with chemical reagents. However, the step of exogenous staining makes these methods undesirable for rapid, nondestructive, and long-term investigation. Here, we present an instantaneous viability assessment of unlabeled cells using phase imaging with computation specificity. This concept utilizes deep learning techniques to compute viability markers associated with the specimen measured by label-free quantitative phase imaging. Demonstrated on different live cell cultures, the proposed method reports approximately 95% accuracy in identifying live and dead cells. The evolution of the cell dry mass and nucleus area for the labeled and unlabeled populations reveal that the chemical reagents decrease viability. The nondestructive approach presented here may find a broad range of applications, from monitoring the production of biopharmaceuticals to assessing the effectiveness of cancer treatments.

Highlights

Existing approaches to evaluate cell viability involve cell staining with chemical reagents
We demonstrate that rapid viability assay can be conducted in a label-free manner using spatial light interference microscopy (SLIM)[35,36], a highly sensitive Quantitative phase imaging (QPI) method, and deep learning
These findings suggest that the phase imaging with computational specificity (PICS) method enables rapid, nondestructive, and unbiased cell viability assessment, potentially valuable to a broad range of biomedical problems, from drug testing to the production of biopharmaceuticals

Summary

Introduction

Existing approaches to evaluate cell viability involve cell staining with chemical reagents. We present an instantaneous viability assessment of unlabeled cells using phase imaging with computation specificity. This concept utilizes deep learning techniques to compute viability markers associated with the specimen measured by label-free quantitative phase imaging. In 2018, Google presented “in silico labeling”, a deep learning based approach that can predict fluorescent labels from transmitted-light (bright field and phase contrast) images of unlabeled samples[23]. By tracking the cell morphology over time, unstained HeLa cells show significantly higher viability compared to the cells stained with viability reagents These findings suggest that the PICS method enables rapid, nondestructive, and unbiased cell viability assessment, potentially valuable to a broad range of biomedical problems, from drug testing to the production of biopharmaceuticals

Methods

Results

Conclusion