Weakly Supervised Deep Learning for Detecting and Counting Dead Cells in Microscopy Images

Chen, Siteng; Li, Ao; Lasick, Kathleen A.; Huynh, Julie M.; Powers, Linda S.; Roveda, Janet; Paek, Andrew L.

doi:10.1109/icmla.2019.00282

“…In any case, there is need for time and money in order to perform these assays. An interesting approach by Chen and collaborators using weakly supervised CNN models demonstrated that they could confidently detect and count dead cells in brightfield images of cell cultures [11].…”

Section: Introductionmentioning

confidence: 99%

celldeath: A tool for detection of cell death in transmitted light microscopy images by deep learning-based visual recognition

Greca

¹

,

Pérez

²

,

Castañeda

³

et al. 2021

View full text Add to dashboard Cite

Cell death experiments are routinely done in many labs around the world, these experiments are the backbone of many assays for drug development. Cell death detection is usually performed in many ways, and requires time and reagents. However, cell death is preceded by slight morphological changes in cell shape and texture. In this paper, we trained a neural network to classify cells undergoing cell death. We found that the network was able to highly predict cell death after one hour of exposure to camptothecin. Moreover, this prediction largely outperforms human ability. Finally, we provide a simple python tool that can broadly be used to detect cell death.

show abstract

“…In any case, there is need for time and money in order to perform these assays. An interesting approach by Chen and collaborators using weakly supervised CNN models demonstrated that they could confidently detect and count dead cells in brightfield images of cell cultures (Chen et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

celldeath: a tool for detection of cell death in transmitted light microscopy images by deep learning-based visual recognition

Greca

¹

,

Pérez

²

,

Milone

³

et al. 2020

Preprint

View full text Add to dashboard Cite

Cell death experiments are routinely done in many labs around the world, these experiments are the backbone of many assays for drug development. Cell death detection is usually performed in many ways, and requires time and reagents. However, cell death is preceded by slight morphological changes in cell shape and texture. In this paper, we trained a neural network to classify cells undergoing cell death. We found that the network was able to highly predict cell death after one hour of exposure to camptothecin. Moreover, this prediction largely outperforms human ability. Finally, we provide a simple python tool that can broadly be used to detect cell death.

show abstract

“…Indeed, computational methods have shown to be more precise, and more rigorous compared to visual/manual annotations 59 , in almost any objective task at hand, e.g. distinguishing living/dead cells 60 , annotating cells, identifying B cells in tissue sections 61 , etc. In this regard, Deep Learning (DL) methods stand out above the rest of the algorithmic approaches in almost any automatization task.…”

Section: Single-cell Analysis (Sca)mentioning

confidence: 99%

Development and validation of image processing and analysis tools for the quantification of the tumor microenvironment in multiplex immunostained cancer tissues.

Jiménez¹

0

View full text Add to dashboard Cite

I would like to express my gratitude to all the people that guided me for the past four years. First and foremost, I would like to thank Carlos Ortiz de Solórzano for allowing me to pursue a PhD under his supervision, from whom I have learned everything I know about science and academia. I would like to give special thanks to him for trusting me through these years, keeping an open mind, giving me the freedom to pursue a variety of research directions, and just simply, for treating me well. Moreover, I am also very grateful to the most experienced lab members: Mikel Ariz, for being such an excellent role model on scientific integrity, and for encouraging me always to try new ideas; Iván Cortés, for being always there, and for teaching me that with hard work and determination everything is possible; and Xabier Morales, the biochemist of the lab, for teaching me a wide range of subjects (from molecular biology to beer fermentation) with great enthusiasm.Joining the Preclinical Models and Analysis Tools research group at CIMA in 2017, has been a wonderful and truly enriching experience. I was coming from another job as a software engineer at a consulting company, and I discovered, quite surprisingly, very passionate people dedicated to cancer research. I have to thank the B.

show abstract

Weakly Supervised Deep Learning for Detecting and Counting Dead Cells in Microscopy Images

Cited by 4 publications

References 19 publications

celldeath: A tool for detection of cell death in transmitted light microscopy images by deep learning-based visual recognition

celldeath: A tool for detection of cell death in transmitted light microscopy images by deep learning-based visual recognition

celldeath: a tool for detection of cell death in transmitted light microscopy images by deep learning-based visual recognition

Development and validation of image processing and analysis tools for the quantification of the tumor microenvironment in multiplex immunostained cancer tissues.

Contact Info

Product

Resources

About