Classification of T cell metabolism from autofluorescence imaging features

Hu, Linghao; Wang, Nianchao; Walsh, A.

doi:10.1117/12.2577004

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning models are appropriate for analysis of datasets with multiple variables, and have been used to extract and interpret cell phenotypes from fluorescence lifetime data. Several studies have applied extracted lifetime features with machine learning algorithms to identify mouse embryo health, quantify precancer cells, classify T cell activation, differentiate stem cell phenotypes, and investigate metabolic perturbations (35,54,(62)(63)(64)(65)(66). Here, both conventional machine learning methods and neural networks were trained with autofluorescence lifetime images and features to predict metabolic states of cancer cells.…”

Section: Discussionmentioning

confidence: 99%

Machine learning prediction of cancer cell metabolism from autofluorescence lifetime images

Wang

Bryant

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Metabolic reprogramming at a cellular level contributes to many diseases including cancer, yet few assays are capable of measuring metabolic pathway usage by individual cells within living samples. Here, we combine autofluorescence lifetime imaging with single cell segmentation and machine-learning models to predict the metabolic pathway usage of cancer cells. The metabolic activities of MCF7 breast cancer cells and HepG2 liver cancer cells were controlled by growing the cells in culture media with specific substrates and metabolic inhibitors. Fluorescence lifetime images of two endogenous metabolic coenzymes, reduced nicotinamide adenine dinucleotide (NADH) and oxidized flavin adenine dinucleotide (FAD), were acquired by a multi-photon fluorescence lifetime microscope and analyzed at the cellular level. Quantitative changes of NADH and FAD lifetime components were observed for cells using glycolysis, oxidative phosphorylation, and glutaminolysis. Conventional machine learning models trained with the autofluorescence features classified cells as dependent on glycolytic or oxidative metabolism with 90 to 92% accuracy. Furthermore, adapting convolutional neural networks to predict cancer cell metabolic perturbations from the autofluorescence lifetime images provided improved performance, 95% accuracy, over traditional models trained via extracted features. In summary, autofluorescence lifetime imaging combined with machine learning models can detect metabolic perturbations between glycolysis and oxidative metabolism of living samples at a cellular level, providing a label free technology to study cellular metabolism and metabolic heterogeneity.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine learning prediction of cancer cell metabolism from autofluorescence lifetime images

Wang

Bryant

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Previously, subcellular features like mitochondrial clusters and networks in the autofluorescence images have been utilized to discriminate different metabolic activities in skin and tumor cells [9,12,17]. Moreover, spatial localization of fluorescence lifetime features including free NAD(P)H fraction (α 1 ), bound FAD fraction (α 1 ), and redox ratios (FAD/(FAD + NAD(P)H)) provides additional metabolic information in T cells [18]. However, the identification of subcellular features in autofluorescence images requires domain expertise and customized data and image processing algorithms.…”

Section: Resultsmentioning

confidence: 99%

Identifying cancer cell metabolic states in autofluorescence lifetime images with machine learning

Walsh

2023

Label-Free Biomedical Imaging and Sensing (LBIS) 2023

View full text Add to dashboard Cite

Cancer cells switch to glycolytic metabolic states even in aerobic environments to support enhanced growth and cellular functions. This phenomenon is known as the Warburg effect, and it inspires advancing interests in targeting metabolism for cancer therapy. Optical metabolic imaging (OMI) measures the fluorescence intensity and lifetime of the co-enzymes reduced nicotinamide adenine dinucleotide (NADH) and oxidized flavin adenine dinucleotide (FAD). OMI can quantitatively distinguish cellular metabolic activities in a label-free manner. The goal of this study is to identify key metabolic pathways of cancer cells using NADH and FAD fluorescence lifetime imaging and machine learning methods. MCF-7 breast cancer cells were exposed to different culture media and inhibitors to disturb their metabolic activities, and NADH and FAD fluorescence lifetime imaging were performed by a multi-photon microscope. Here, we proposed a potential method of training convolutional neural networks to predict cellular metabolic states. Adapting convolutional neural networks for the prediction of cancer cell metabolic conditions was anticipated to provide substantially better performance than traditional models with extracted features. In summary, this investigation offers a non-invasive, quantitative technology to detect metabolic perturbations at a cellular level, which improves the identification of different metabolic states of cancer cells.

show abstract

“…Changes in metabolic pathways within MCF7 cells alter the fluorescence lifetimes of NAD(P)H and FAD. The reduction of free NAD(P)H fraction (α 1 ) with 2-DG treatment was observed in MCF10A breast cancer cells and pancreatic islet cells (Drozdowicz-Tomsia et al, 2014;Wang et al, 2021). Conversely, a higher level of glycolysis induced more fraction of free NAD(P)H in kidney cells, neural cells, and stem cells in disease states (Stringari et al, 2012a;Chakraborty et al, 2016;Sameni et al, 2016).…”

Section: Discussionmentioning

confidence: 98%

“…Machine learning models are appropriate for analysis of datasets with multiple variables, and have been used to extract and interpret cell phenotypes from fluorescence lifetime data. Several studies have applied extracted lifetime features with machine learning algorithms to identify mouse embryo health, quantify precancer cells, classify T cell activation, differentiate stem cell phenotypes, and investigate metabolic perturbations ( Gu et al, 2015 ; Liu et al, 2018 ; Ma et al, 2019 ; Wang et al, 2020 ; Hu et al, 2021 ; Qian et al, 2021 ; Walsh et al, 2021 ). Here, both conventional machine learning methods and neural networks were trained with autofluorescence lifetime images and features to predict metabolic states of cancer cells.…”

Section: Discussionmentioning

confidence: 99%

Label-free spatially maintained measurements of metabolic phenotypes in cells

Hu,

Wang,

Bryant

et al. 2023

Front. Bioeng. Biotechnol.

Self Cite

View full text Add to dashboard Cite

Metabolic reprogramming at a cellular level contributes to many diseases including cancer, yet few assays are capable of measuring metabolic pathway usage by individual cells within living samples. Here, autofluorescence lifetime imaging is combined with single-cell segmentation and machine-learning models to predict the metabolic pathway usage of cancer cells. The metabolic activities of MCF7 breast cancer cells and HepG2 liver cancer cells were controlled by growing the cells in culture media with specific substrates and metabolic inhibitors. Fluorescence lifetime images of two endogenous metabolic coenzymes, reduced nicotinamide adenine dinucleotide (NADH) and oxidized flavin adenine dinucleotide (FAD), were acquired by a multi-photon fluorescence lifetime microscope and analyzed at the cellular level. Quantitative changes of NADH and FAD lifetime components were observed for cells using glycolysis, oxidative phosphorylation, and glutaminolysis. Conventional machine learning models trained with the autofluorescence features classified cells as dependent on glycolytic or oxidative metabolism with 90%–92% accuracy. Furthermore, adapting convolutional neural networks to predict cancer cell metabolic perturbations from the autofluorescence lifetime images provided improved performance, 95% accuracy, over traditional models trained via extracted features. Additionally, the model trained with the lifetime features of cancer cells could be transferred to autofluorescence lifetime images of T cells, with a prediction that 80% of activated T cells were glycolytic, and 97% of quiescent T cells were oxidative. In summary, autofluorescence lifetime imaging combined with machine learning models can detect metabolic perturbations between glycolysis and oxidative metabolism of living samples at a cellular level, providing a label-free technology to study cellular metabolism and metabolic heterogeneity.

show abstract

Classification of T cell metabolism from autofluorescence imaging features

Cited by 3 publications

References 0 publications

Machine learning prediction of cancer cell metabolism from autofluorescence lifetime images

Machine learning prediction of cancer cell metabolism from autofluorescence lifetime images

Identifying cancer cell metabolic states in autofluorescence lifetime images with machine learning

Label-free spatially maintained measurements of metabolic phenotypes in cells

Contact Info

Product

Resources

About