Stem cell motion-tracking by using deep neural networks with multi-output

Yang-xu, Wang; Mao, Hua; Zhang, Yi

doi:10.1007/s00521-017-3291-2

Cited by 30 publications

(29 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, non-invasive tools to assess the in vitro viability, metabolic activity and structures in real time are necessary to develop in regenerative medicine. Recent advances in optics and imaging analysis, such as time-lapse microscopic images combined with computational machine learning 4 , 37 , 38 and the use of a specific sensor 39 , have provided many opportunities to address these issues for the QC of cells to be engineered. Previously, we applied two image analysis algorithms of OF and NCC to p0 oral keratinocytes during the early stage of 2D cell culture and confirmed that the two motion indices calculated by both algorithms were applicable and feasible for non-invasive monitoring of cells and can be used as an appropriate tool for QC 29 .…”

Section: Discussionmentioning

confidence: 99%

Cells/colony motion of oral keratinocytes determined by non-invasive and quantitative measurement using optical flow predicts epithelial regenerative capacity

Hoshikawa

Sato

Haga

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Cells/colony motion determined by non-invasive, quantitative measurements using the optical flow (OF) algorithm can indicate the oral keratinocyte proliferative capacity in early-phase primary cultures. This study aimed to determine a threshold for the cells/colony motion index to detect substandard cell populations in a subsequent subculture before manufacturing a tissue-engineered oral mucosa graft and to investigate the correlation with the epithelial regenerative capacity. The distinctive proliferating pattern of first-passage [passage 1 (p1)] cells reveals the motion of p1 cells/colonies, which can be measured in a non-invasive, quantitative manner using OF with fewer full-screen imaging analyses and cell segmentations. Our results demonstrate that the motion index lower than 40 μm/h reflects cellular damages by experimental metabolic challenges although this value shall only apply in case of our culture system. Nonetheless, the motion index can be used as the threshold to determine the quality of cultured cells while it may be affected by any different culture conditions. Because the p1 cells/colony motion index is correlated with epithelial regenerative capacity, it is a reliable index for quality control of oral keratinocytes.

show abstract

Section: Discussionmentioning

confidence: 99%

Cells/colony motion of oral keratinocytes determined by non-invasive and quantitative measurement using optical flow predicts epithelial regenerative capacity

Hoshikawa

Sato

Haga

et al. 2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…CV in microscopy: Advances in microscopes and computational hardware are expanding the possibilities for live-cell image analysis, which is of importance to research in biology. Deep neural networks [50,55] are used for tracking of cells or simulated particles. Detection based tracking [49] and feature tracking [36,40] were successfully applied in cell migration analysis [26].…”

Section: Related Workmentioning

confidence: 99%

Learning Nanoscale Motion Patterns of Vesicles in Living Cells

Sekh

Opstad

Birgisdottir

et al. 2020

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Detecting and analyzing nanoscale motion patterns of vesicles, smaller than the microscope resolution (∼ 250 nm), inside living biological cells is a challenging problem. Stateof-the-art CV approaches based on detection, tracking, optical flow or deep learning perform poorly on this problem. We propose an integrative approach built upon physics-based simulations, nanoscopy algorithms and shallow residual attention network to permit for the first time analysis of subresolution motion patterns in vesicles, also of sub-resolution diameter. Our results show state-of-the-art performance, 89% validation accuracy on simulated dataset and 82% testing accuracy on an experimental dataset of images of living heart muscle cells grown under three different pathophysiologically relevant conditions. We demonstrate automated analysis of the motion states and changes in them for over 9000 vesicles. Such analysis will enable large scale biological studies of vesicle transport and interactions in living cells in the future.

show abstract

“…Several classes of convolutional neural nets (CNNs) have been developed specifically to perform dense cell segmentation (Xie et al, 2016), based upon different architectures such as autoencoders (Su et al, 2015), U-Nets (Falk et al, 2019;Ronneberger et al, 2015;Xie et al, 2018), or variants of the Inception architecture (Cohen et al, 2017;Szegedy et al, 2014). Recent approaches have further enhanced cell-tracking fidelity by combining deep cell bounding-box detection with ancillary tasks such as morphology classification (He et al, 2017) or mitosis detection (Wang et al, 2019), improving accuracy, although requiring additional annotation. Further, preprocessing and filtering steps to rapidly reject irrelevant images (Araú jo et al, 2019) have enabled highly scalable deep cell segmentation pipelines.…”

Section: Introductionmentioning

confidence: 99%

“…Further, preprocessing and filtering steps to rapidly reject irrelevant images (Araú jo et al, 2019) have enabled highly scalable deep cell segmentation pipelines. Applying adaptive tracking algorithms such as particle filters (He et al, 2017;Wang et al, 2019) has enabled cell tracking under conditions of high morphological diversity and low image resolution, although the impact of cell density on tracking fidelity was not examined. In contrast, Hiramatsu et al (2018) developed a novel pipeline where individual U-Net CNNs, each trained independently, were combined using a gating network to produce a high-quality, dense cell segmentation, although the impact of neural net architecture on the final result was not explored.…”

Section: Introductionmentioning

confidence: 99%

Deep neural net tracking of human pluripotent stem cells reveals intrinsic behaviors directing morphogenesis

2021

View full text Add to dashboard Cite

Lineage tracing is a powerful tool in developmental biology to interrogate the evolution of tissue formation, but the dense, three-dimensional nature of tissue limits the assembly of individual cell trajectories into complete reconstructions of development. Human induced pluripotent stem cells (hiPSCs) can recapitulate aspects of developmental processes, providing an in vitro platform to assess the dynamic collective behaviors directing tissue morphogenesis. Here, we trained an ensemble of neural networks to track individual hiPSCs in timelapse microscopy, generating longitudinal measures of cell and cellular neighborhood properties on timescales from minutes to days. Our analysis reveals that, while individual cell parameters are not strongly affected by pluripotency maintenance conditions or morphogenic cues, regional changes in cell behavior predict cell fate and colony organization. By generating complete multicellular reconstructions of hiPSC behavior, our tracking pipeline enables fine-grained understanding of morphogenesis by elucidating the role of regional behavior in early tissue formation.

show abstract

Stem cell motion-tracking by using deep neural networks with multi-output

Cited by 30 publications

References 25 publications

Cells/colony motion of oral keratinocytes determined by non-invasive and quantitative measurement using optical flow predicts epithelial regenerative capacity

Cells/colony motion of oral keratinocytes determined by non-invasive and quantitative measurement using optical flow predicts epithelial regenerative capacity

Learning Nanoscale Motion Patterns of Vesicles in Living Cells

Deep neural net tracking of human pluripotent stem cells reveals intrinsic behaviors directing morphogenesis

Contact Info

Product

Resources

About