Lcuts: Linear Clustering of Bacteria Using Recursive Graph Cuts

Wang

et al. 2020

Nat Commun

Fluorescence microscopy enables spatial and temporal measurements of live cells and cellular communities. However, this potential has not yet been fully realized for investigations of individual cell behaviors and phenotypic changes in dense, three-dimensional (3D) bacterial biofilms. Accurate cell detection and cellular shape measurement in densely packed biofilms are challenging because of the limited resolution and low signal to background ratios (SBRs) in fluorescence microscopy images. In this work, we present Bacterial Cell Morphometry 3D (BCM3D), an image analysis workflow that combines deep learning with mathematical image analysis to accurately segment and classify single bacterial cells in 3D fluorescence images. In BCM3D, deep convolutional neural networks (CNNs) are trained using simulated biofilm images with experimentally realistic SBRs, cell densities, labeling methods, and cell shapes. We systematically evaluate the segmentation accuracy of BCM3D using both simulated and experimental images. Compared to state-of-the-art bacterial cell segmentation approaches, BCM3D consistently achieves higher segmentation accuracy and further enables automated morphometric cell classifications in multi-population biofilms.

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Cell Segmentation By Thresholding Cnn Confidence Mapsmentioning

confidence: 99%

See 1 more Smart Citation

Non-invasive single-cell morphometry in living bacterial biofilms

Wang

et al. 2020

Nat Commun

“…These results show that, when using BCM3D 2.0, ~86% of cells are segmented with physiologically reasonable cell shapes. The remaining 14% of cells can then either be excluded from the analyses or be to further processing to identify and correct the remaining segmentation errors 23,24,49 .…”

Section: Segmentation Of Experimentally Obtained Biofilm Imagesmentioning

confidence: 99%

BCM3D 2.0: Accurate segmentation of single bacterial cells in dense biofilms using computationally generated intermediate image representations

Wang²,

Donarski³

et al. 2021

Preprint

Accurate detection and segmentation of single cells in three-dimensional (3D) fluorescence time-lapse images is essential for measuring individual cell behaviors in large bacterial communities called biofilms. Recent progress in machine-learning-based image analysis is providing this capability with every increasing accuracy. Leveraging the capabilities of deep convolutional neural networks (CNNs), we recently developed bacterial cell morphometry in 3D (BCM3D), an integrated image analysis pipeline that combines deep learning with conventional image analysis to detect and segment single biofilm-dwelling cells in 3D fluorescence images. While the first release of BCM3D (BCM3D 1.0) achieved state-of-the-art 3D bacterial cell segmentation accuracies, low signal-to-background ratios (SBRs) and images of very dense biofilms remained challenging. Here, we present BCM3D 2.0 to address this challenge. BCM3D 2.0 is completely complementary to the approach utilized in BCM3D 1.0. Instead of training CNNs to perform voxel classification, we trained CNNs to translate 3D fluorescence images into intermediate 3D image representations that are, when combined appropriately later, more amenable to conventional mathematical image processing than a single experimental image. Using this approach, improved segmentation results are obtained even for very low SBRs and/or high cell density biofilm images. The improved cell segmentation accuracies in turn enable improved accuracies of tracking individual cells through 3D space and time, which opens the door to investigating time-dependent phenomena in bacterial biofilms at the cellular level.

“…Individual cells in biofilms interact 38 with other cells, the ECM, or with the substrate surface, and the sum total of these interactions 39 provide bacterial biofilms with emergent functional capabilities beyond those of individual cells. 40 For example, biofilms are orders of magnitude more tolerant towards physical, chemical, and 41 biological stressors, including antibiotic treatments and immune system clearance 1,2,[5][6][7][8] . 42 Understanding how such capabilities emerge from the coordination of individual cell behavior 43 requires imaging technologies capable of resolving and simultaneous tracking of individual 44 bacterial cells in 3D biofilms.…”

mentioning

confidence: 99%

Non-Invasive Single-Cell Morphometry in Living Bacterial Biofilms

Wang

et al. 2020

Preprint

20Fluorescence microscopy enables spatial and temporal measurements of live cells and cellular 21 communities. However, this potential has not yet been fully realized for investigations of 22 individual cell behaviors and phenotypic changes in dense, three-dimensional (3D) bacterial 23 biofilms. Accurate cell detection and cellular shape measurement in densely packed biofilms are 24 challenging because of the limited resolution and low signal to background ratio in fluorescence 25 microscopy images. In this work, we present Bacterial Cell Morphometry 3D (BCM3D), an image 26 analysis workflow that combines deep learning with mathematical image analysis to accurately 27 segment and classify single bacterial cells in 3D fluorescence images. In BCM3D, deep 28 convolutional neural networks (CNNs) are trained using simulated biofilm images with 29 experimentally realistic signal-to-background ratios (SBRs), cell densities, labelling methods, and 30 cell shapes. We systematically evaluated the segmentation accuracy of BCM3D using both 31 simulated and experimental images. Compared to state-of-the-art bacterial cell segmentation 32 approaches, BCM3D consistently achieves higher segmentation accuracy and further enables 33 automated morphometric cell classifications in multi-population biofilms.Biofilms are multicellular communities of microorganisms that grow on biotic or abiotic 36 surfaces 1-4 . In addition to cellular biomass, biofilms also contain an extracellular matrix (ECM) 37 which is composed of polysaccharides, DNA, and proteins. Individual cells in biofilms interact 38 with other cells, the ECM, or with the substrate surface, and the sum total of these interactions 39 provide bacterial biofilms with emergent functional capabilities beyond those of individual cells. 40 For example, biofilms are orders of magnitude more tolerant towards physical, chemical, and 41 biological stressors, including antibiotic treatments and immune system clearance 1,2,[5][6][7][8] . 42 Understanding how such capabilities emerge from the coordination of individual cell behavior 43 requires imaging technologies capable of resolving and simultaneous tracking of individual 44 bacterial cells in 3D biofilms. 45 Live cell-compatible imaging technologies, such as light microscopy, can reveal the spatial 46 and temporal context that affects cellular behaviors. However, conventional imaging modalities 47 are not able to resolve individual cells within thick 3D biofilms over extended periods of time. For 48 example, the diffraction-limited lateral x,y-resolution (~230 nm) of a confocal fluorescence 49 microscope is barely sufficient to resolve bacterial cells positioned next to each other on flat glass 50 coverslips. Even worse, the diffraction-limited axial z-resolution (570 nm) is comparable to the 51 size of a single bacterial cell, so that densely-packed cells become unresolvable in the axial z-52 dimension 9,10 . Notable exceptions include loose biofilms (low cell density), spherical cell 53 shapes 11,12 , and mutan...