Complementary performances of convolutional and capsule neural networks on classifying microfluidic images of dividing yeast cells

Abstract

Microfluidic-based assays have become effective high-throughput approaches to examining replicative aging of budding yeast cells. Deep learning may offer an efficient way to analyze a large number of images collected from microfluidic experiments. Here, we compare three deep learning architectures to classify microfluidic time-lapse images of dividing yeast cells into categories that represent different stages in the yeast replicative aging process. We found that convolutional neural networks outperformed capsule networks in terms of accuracy, precision, and recall. The capsule networks had the most robust performance in detecting one specific category of cell images. An ensemble of three best-fitted single-architecture models achieves the highest overall accuracy, precision, and recall due to complementary performances. In addition, extending classification classes and data augmentation of the training dataset can improve the predictions of the biological categories in our study. This work lays a useful framework for sophisticated deep-learning processing of microfluidic-based assays of yeast replicative aging.

The Yeast Lifespan Machine: a microfluidic platform for automated replicative lifespan measurements

“…We note that a number of other advances in microfluidic technologies and automated image analysis have been applied to this problem and reported by other groups (17, 27, 28). While it is difficult to directly compare these methods, various improvements in microfluidic layout, cell retention features, image collection, and imaging analysis could be combined to create more robust and accessible tools for measuring the lifespan of yeast cells.…”

“…Combined with the improvements to image analysis, these advances allow us to measure the lifespan under multiple genetic and environmental conditions, all with minimal human setup, intervention, or annotation of the resulting raw data. We note that a number of other advances in microfluidic technologies and automated image analysis have been applied to this problem and reported by other groups (17,27,28). While it is difficult to directly compare these methods, various improvements in microfluidic layout, cell retention features, image collection, and imaging analysis could be combined to create more robust and accessible tools for measuring the lifespan of yeast cells.…”

DetecDiv, a generalist deep-learning platform for automated cell division tracking and survival analysis

“…The initial development of microfluidic systems for RLS assays has partially alleviated this problem by allowing continuous observation of individual cell divisions and relevant fluorescent cellular markers under the microscope from birth to death (Lee et al 2012; Xie et al 2012; Fehrmann et al 2013). Recent efforts further increased data acquisition throughput (Jo et al 2015; Liu, Young, and Acar 2015) and attempted to automate data analysis (Ghafari et al 2021; Ghafari, Mailman, and Qin 2021). Yet, retrieving individual cellular lifespans from large sets of image sequences so far remained an insurmountable bottleneck to characterize senescence entry quantitatively or to screen large numbers of mutants and environmental conditions.…”

“…Recently, a study showed the potential of image classification by a CNN or a capsule network to classify the state of dividing yeast cells (i.e., budded, unbudded, etc. ) trapped in a microfluidic device (Ghafari et al 2021). However, due to the limited accuracy of the model, it has not demonstrated its ability to perform an automated division counting, let alone determine the RLS of individual cells.…”

“…e ., budded, unbudded, etc.) trapped in a microfluidic device (Ghafari et al 2021). However, due to the limited accuracy of the model, it has not demonstrated its ability to perform an automated division counting, let alone determine the RLS of individual cells.…”