scholarly journals MiSiC, a general deep learning-based method for the high-throughput cell segmentation of complex bacterial communities

eLife ◽  
2021 ◽  
Vol 10 ◽  
Author(s):  
Swapnesh Panigrahi ◽  
Dorothée Murat ◽  
Antoine Le Gall ◽  
Eugénie Martineau ◽  
Kelly Goldlust ◽  
...  
Keyword(s):  
Deep Learning ◽  
Bacterial Communities ◽  
Cell Biology ◽  
Imaging Modality ◽  
Interaction Model ◽  
Cell Segmentation ◽  
Bacterial Cells ◽  
Interspecies Interactions ◽  
Prey Interaction ◽  
Simple Implementation

Studies of bacterial communities, biofilms and microbiomes, are multiplying due to their impact on health and ecology. Live imaging of microbial communities requires new tools for the robust identification of bacterial cells in dense and often inter-species populations, sometimes over very large scales. Here, we developed MiSiC, a general deep-learning-based 2D segmentation method that automatically segments single bacteria in complex images of interacting bacterial communities with very little parameter adjustment, independent of the microscopy settings and imaging modality. Using a bacterial predator-prey interaction model, we demonstrate that MiSiC enables the analysis of interspecies interactions, resolving processes at subcellular scales and discriminating between species in millimeter size datasets. The simple implementation of MiSiC and the relatively low need in computing power make its use broadly accessible to fields interested in bacterial interactions and cell biology.

scholarly journals MiSiC, a general deep learning-based method for the high-throughput cell segmentation of complex bacterial communities

2020 ◽  
Author(s):  
Swapnesh Panigrahi ◽  
Dorothée Murat ◽  
Antoine Le Gall ◽  
Eugénie Martineau ◽  
Kelly Goldlust ◽  
...  
Keyword(s):  
Deep Learning ◽  
Bacterial Communities ◽  
Cell Biology ◽  
Imaging Modality ◽  
Interaction Model ◽  
Cell Segmentation ◽  
Bacterial Cells ◽  
Interspecies Interactions ◽  
Wide Range ◽  
Prey Interaction

AbstractStudies of microbial communities by live imaging require new tools for the robust identification of bacterial cells in dense and often inter-species populations, sometimes over very large scales. Here, we developed MiSiC, a general deep-learning-based segmentation method that automatically segments a wide range of spatially structured bacterial communities with very little parameter adjustment, independent of the imaging modality. Using a bacterial predator-prey interaction model, we demonstrate that MiSiC enables the analysis of interspecies interactions, resolving processes at subcellular scales and discriminating between species in millimeter size datasets. The simple implementation of MiSiC and the relatively low need in computing power make its use broadly accessible to fields interested in bacterial interactions and cell biology.

scholarly journals Scellseg: a style-aware cell instance segmentation tool with pre-training and contrastive fine-tuning

2021 ◽  
Author(s):  
Dejin Xun ◽  
Deheng Chen ◽  
Yitian Zhou ◽  
Volker M. Lauschke ◽  
Rui Wang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Network Architecture ◽  
Cell Biology ◽  
Large Scale ◽  
Training Data ◽  
Fine Tuning ◽  
Cell Segmentation ◽  
Model Tuning ◽  
Scale Experiment ◽  
Tuning Strategy

Deep learning-based cell segmentation is increasingly utilized in cell biology and molecular pathology, due to massive accumulation of diverse large-scale datasets and excellent performance in cell representation. However, the development of specialized algorithms has long been hampered by a paucity of annotated training data, whereas the performance of generalist algorithm was limited without experiment-specific calibration. Here, we present a deep learning-based tool called Scellseg consisted of novel pre-trained network architecture and contrastive fine-tuning strategy. In comparison to four commonly used algorithms, Scellseg outperformed in average precision on three diverse datasets with no need for dataset-specific configuration. Interestingly, we found that eight images are sufficient for model tuning to achieve satisfied performance based on a shot data scale experiment. We also developed a graphical user interface integrated with functions of annotation, fine-tuning and inference, that allows biologists to easily specialize their own segmentation model and analyze data at the single-cell level.

scholarly journals BactMAP: an R package for integrating, analyzing and visualizing bacterial microscopy data

2019 ◽  
Author(s):  
Renske van Raaphorst ◽  
Morten Kjos ◽  
Jan-Willem Veening
Keyword(s):  
Cell Biology ◽  
R Package ◽  
Cell Segmentation ◽  
Bacterial Cells ◽  
Detection Algorithms ◽  
Spot Detection ◽  
Z Ring ◽  
Microscopy Data ◽  
Cell Microscopy ◽  
Different Sources

AbstractHigh-throughput analyses of single-cell microscopy data is a critical tool within the field of bacterial cell biology. Several programs have been developed to specifically segment bacterial cells from phase-contrast images. Together with spot and object detection algorithms, these programs offer powerful approaches to quantify observations from microscopy data, ranging from cell-to-cell genealogy to localization and movement of proteins. Most segmentation programs contain specific post-processing and plotting options, but these options vary between programs and possibilities to optimize or alter the outputs are often limited. Therefore, we developed BactMAP (Bacterial toolbox for Microscopy Analysis & Plotting), a software package that allows researchers to transform cell segmentation and spot detection data generated by different programs automatically into various plots. Furthermore, BactMAP makes it possible to perform custom analyses and change the layout of the output. Because BactMAP works independently of segmentation and detection programs, inputs from different sources can be compared within the same analysis pipeline. BactMAP runs in R, which enables the use of advanced statistical analysis tools as well as easily adjustable plot graphics in every operating system. Using BactMAP we visualize key cell cycle parameters in Bacillus subtilis and Staphylococcus aureus, and demonstrate that the DNA replication forks in Streptococcus pneumoniae dissociate and associate before splitting of the cell, after the Z-ring is formed at the new quarter positions. BactMAP is available from https://veeninglab.com/bactmap.

scholarly journals Synthetic polarization-sensitive optical coherence tomography by deep learning

2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Yi Sun ◽  
Jianfeng Wang ◽  
Jindou Shi ◽  
Stephen A. Boppart
Keyword(s):  
Optical Coherence Tomography ◽  
Deep Learning ◽  
Imaging System ◽  
Imaging Modality ◽  
Similarity Index ◽  
Polarization Sensitivity ◽  
Imaging Method ◽  
Optical Coherence ◽  
The Real ◽  
Oct Imaging

AbstractPolarization-sensitive optical coherence tomography (PS-OCT) is a high-resolution label-free optical biomedical imaging modality that is sensitive to the microstructural architecture in tissue that gives rise to form birefringence, such as collagen or muscle fibers. To enable polarization sensitivity in an OCT system, however, requires additional hardware and complexity. We developed a deep-learning method to synthesize PS-OCT images by training a generative adversarial network (GAN) on OCT intensity and PS-OCT images. The synthesis accuracy was first evaluated by the structural similarity index (SSIM) between the synthetic and real PS-OCT images. Furthermore, the effectiveness of the computational PS-OCT images was validated by separately training two image classifiers using the real and synthetic PS-OCT images for cancer/normal classification. The similar classification results of the two trained classifiers demonstrate that the predicted PS-OCT images can be potentially used interchangeably in cancer diagnosis applications. In addition, we applied the trained GAN models on OCT images collected from a separate OCT imaging system, and the synthetic PS-OCT images correlate well with the real PS-OCT image collected from the same sample sites using the PS-OCT imaging system. This computational PS-OCT imaging method has the potential to reduce the cost, complexity, and need for hardware-based PS-OCT imaging systems.

scholarly journals Efficient Cell Segmentation from Electroluminescent Images of Single-Crystalline Silicon Photovoltaic Modules and Cell-Based Defect Identification Using Deep Learning with Pseudo-Colorization

Sensors ◽  
2021 ◽  
Vol 21 (13) ◽  
pp. 4292
Author(s):  
Horng-Horng Lin ◽  
Harshad Kumar Dandage ◽  
Keh-Moh Lin ◽  
You-Teh Lin ◽  
Yeou-Jiunn Chen
Keyword(s):  
Deep Learning ◽  
Defect Detection ◽  
Crystalline Silicon ◽  
Cell Segmentation ◽  
Solar Pv ◽  
Single Crystalline Silicon ◽  
Defect Inspection ◽  
Defect Identification ◽  
Pv Modules ◽  
A Cell

Solar cells may possess defects during the manufacturing process in photovoltaic (PV) industries. To precisely evaluate the effectiveness of solar PV modules, manufacturing defects are required to be identified. Conventional defect inspection in industries mainly depends on manual defect inspection by highly skilled inspectors, which may still give inconsistent, subjective identification results. In order to automatize the visual defect inspection process, an automatic cell segmentation technique and a convolutional neural network (CNN)-based defect detection system with pseudo-colorization of defects is designed in this paper. High-resolution Electroluminescence (EL) images of single-crystalline silicon (sc-Si) solar PV modules are used in our study for the detection of defects and their quality inspection. Firstly, an automatic cell segmentation methodology is developed to extract cells from an EL image. Secondly, defect detection can be actualized by CNN-based defect detector and can be visualized with pseudo-colors. We used contour tracing to accurately localize the panel region and a probabilistic Hough transform to identify gridlines and busbars on the extracted panel region for cell segmentation. A cell-based defect identification system was developed using state-of-the-art deep learning in CNNs. The detected defects are imposed with pseudo-colors for enhancing defect visualization using K-means clustering. Our automatic cell segmentation methodology can segment cells from an EL image in about 2.71 s. The average segmentation errors along the x-direction and y-direction are only 1.6 pixels and 1.4 pixels, respectively. The defect detection approach on segmented cells achieves 99.8% accuracy. Along with defect detection, the defect regions on a cell are furnished with pseudo-colors to enhance the visualization.

Bacterial presence in polar regions associated with environment modification by chemical compounds including contaminants

2017 ◽  
Vol 25 (4) ◽  
pp. 481-491 ◽  
Author(s):  
Klaudia Kosek ◽  
Katarzyna Jankowska ◽  
Żaneta Polkowska
Keyword(s):  
Bacterial Communities ◽  
Chemical Compounds ◽  
Freshwater Ecosystems ◽  
Trophic Levels ◽  
Polar Regions ◽  
Bacterial Cells ◽  
Chemical Contamination ◽  
Climate Conditions ◽  
Biological Communities ◽  
Long Time

Microbes are omnipresent and diverse members of all biological communities. In marine and freshwater ecosystems, microorganisms form the base of the food chain supporting higher trophic levels. Even though microbes are generally thought to live in warm regions of Earth, many of them develop in cold climates. Polar regions remain relatively protected from widespread anthropogenic disturbances, which is a consequence of thier remoteness and extreme climate conditions. For a long time these regions were considered to be free from chemical contamination until scientists discovered a presence of pollutants there. Chemical contamination may induce serious disorders in the integrity of polar ecosystems influencing the growth of bacterial communities. Xenobiotics including persistent organic pollutants are transported thousands of kilometers by the air and ocean currents, and they are deposed in high-latitude regions and accumulate in all elements of the environment including bacterial communities. It is important to determine their concentration levels in bacterial cells to assess the possibility of contaminants becoming transferred to higher trophic levels; however, some species of bacteria are capable of metabolizing xenobiotics, which makes them less toxic or even removes them from the environment.

scholarly journals A Mask-guided Attention Deep Learning Model for COVID-19 Diagnosis based on an Integrated CT Scan Images Database

2022 ◽  
Author(s):  
Maede Maftouni ◽  
Bo Shen ◽  
Andrew Chung Chee Law ◽  
Niloofar Ayoobi Yazdi ◽  
Zhenyu Kong
Keyword(s):  
Deep Learning ◽  
Ct Scan ◽  
Imaging Modality ◽  
Learning Model ◽  
Classification Performance ◽  
Computer Assisted ◽  
Learning Approach ◽  
Learning Models ◽  
Task Learning ◽  
Data Efficiency

<p>The global extent of COVID-19 mutations and the consequent depletion of hospital resources highlighted the necessity of effective computer-assisted medical diagnosis. COVID-19 detection mediated by deep learning models can help diagnose this highly contagious disease and lower infectivity and mortality rates. Computed tomography (CT) is the preferred imaging modality for building automatic COVID-19 screening and diagnosis models. It is well-known that the training set size significantly impacts the performance and generalization of deep learning models. However, accessing a large dataset of CT scan images from an emerging disease like COVID-19 is challenging. Therefore, data efficiency becomes a significant factor in choosing a learning model. To this end, we present a multi-task learning approach, namely, a mask-guided attention (MGA) classifier, to improve the generalization and data efficiency of COVID-19 classification on lung CT scan images.</p><p>The novelty of this method is compensating for the scarcity of data by employing more supervision with lesion masks, increasing the sensitivity of the model to COVID-19 manifestations, and helping both generalization and classification performance. Our proposed model achieves better overall performance than the single-task baseline and state-of-the-art models, as measured by various popular metrics. In our experiment with different percentages of data from our curated dataset, the classification performance gain from this multi-task learning approach is more significant for the smaller training sizes. Furthermore, experimental results demonstrate that our method enhances the focus on the lesions, as witnessed by both</p><p>attention and attribution maps, resulting in a more interpretable model.</p>

scholarly journals Spatial structure arising from chase-escape interactions with crowding

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Anudeep Surendran ◽  
Michael J. Plank ◽  
Matthew J. Simpson
Keyword(s):  
Spatial Structure ◽  
Cell Biology ◽  
Scale Up ◽  
Interaction Strength ◽  
Distinct Species ◽  
Interspecies Interactions ◽  
Individual Based Model ◽  
Directional Bias ◽  
Spatial Moments ◽  
Individual Level

Abstract Movement of individuals, mediated by localised interactions, plays a key role in numerous processes including cell biology and ecology. In this work, we investigate an individual-based model accounting for various intraspecies and interspecies interactions in a community consisting of two distinct species. In this framework we consider one species to be chasers and the other species to be escapees, and we focus on chase-escape dynamics where the chasers are biased to move towards the escapees, and the escapees are biased to move away from the chasers. This framework allows us to explore how individual-level directional interactions scale up to influence spatial structure at the macroscale. To focus exclusively on the role of motility and directional bias in determining spatial structure, we consider conservative communities where the number of individuals in each species remains constant. To provide additional information about the individual-based model, we also present a mathematically tractable deterministic approximation based on describing the evolution of the spatial moments. We explore how different features of interactions including interaction strength, spatial extent of interaction, and relative density of species influence the formation of the macroscale spatial patterns.

Sign in / Sign up

Export Citation Format

Share Document