Biofilm image reconstruction for assessing structural parameters

The structure of biofilms can be numerically quantified from microscopy images using structural parameters. These parameters are used in biofilm image analysis to compare biofilms, to monitor temporal variation in biofilm structure, to quantify the effects of antibiotics on biofilm structure and to determine the effects of environmental conditions on biofilm structure. It is often hypothesized that biofilms with similar structural parameter values will have similar structures; however, this hypothesis has never been tested. The main goal was to test the hypothesis that the commonly used structural parameters can characterize the differences or similarities between biofilm structures. To achieve this goal (1) biofilm image reconstruction was developed as a new tool for assessing structural parameters, (2) independent reconstructions using the same starting structural parameters were tested to see how they differed from each other, (3) the effect of the original image parameter values on reconstruction success was evaluated, and (4) the effect of the number and type of the parameters on reconstruction success was evaluated. It was found that two biofilms characterized by identical commonly used structural parameter values may look different, that the number and size of clusters in the original biofilm image affect image reconstruction success and that, in general, a small set of arbitrarily selected parameters may not reveal relevant differences between biofilm structures.     

Three-dimensional biofilm image reconstruction for assessing structural parameters

Parameters representing three-dimensional (3D) biofilm structure are quantified from confocal laser-scanning microscope (CLSM) images. These 3D parameters describe the distribution of biomass pixels within the space occupied by a biofilm; however, they lack a direct connection to biofilm activity. As a result, researchers choose a handful of parameters without there being a consensus on a standard set of parameters. We hypothesized that a select 3D parameter set could be used to reconstruct a biofilm image and that the reconstructed and original biofilm images would have similar activities. To test this hypothesis, an algorithm was developed to reconstruct a biofilm image with parameters identical to those of the original CLSM image. We introduced an objective method to assess the reconstruction algorithm by comparing the activities of the original and reconstructed biofilm images. We found that biofilm images with identical structural parameters showed nearly identical activities and substrate concentration profiles. This implies that the set containing all common structural parameters can successfully describe biofilm structure. This finding is significant, as it opens the door to the next step, of finding a smaller standard set of biofilm structural parameters that can be used to compare biofilm structure.     

Investigation of the mesoscale structure and volumetric features of biofilms using optical coherence tomography

Optical coherence tomography (OCT) was successfully applied to visualize the mesoscale structure of three different heterotrophic biofilms. For this purpose, biofilm volumes of 4 × 4 × 1.6 mm³ were scanned with spatial resolutions lower than 20 µm within an acquisition time of 2 min. A heterogeneous structure was detected for biofilms cultivated in laminar as well as transient flow conditions. The structure was found to be more homogeneous for the biofilm grown in turbulent flow. This biofilm structure was characterized by a volumetric porosity of 0.36, whereas the porosity calculated for biofilms grown in laminar and transient conditions was 0.65. These results were directly generated from the distribution of porosity calculated from the OCT images acquired and can be linked to structural properties. Up to now, the mesoscale biofilm structure was only observable with time‐consuming and expensive studies, for example, magnetic resonance microscopy. OCT will most certainly be helpful for improved understanding and prediction of biofilm physics with respect to macroscale processes, for example, mass transfer and detachment as the information about mesoscale is easily accessible using this method. In the context of this study, we show that CLSM images do not necessarily provide an accurate representation of the biofilm structure at the mesoscale. Additionally, the typical characteristic parameters obtained from CLSM image stacks differ largely from those calculated from OCT images. Nevertheless, to determine the local distribution of biofilm constituents, microscopic methods such as confocal laser scanning microscopy are required. Biotechnol. Bioeng. 2010;107: 844–853. © 2010 Wiley Periodicals, Inc.     

Remote Magnetic Actuation of Micrometric Probes for in situ 3D Mapping of Bacterial Biofilm Physical Properties

Bacterial adhesion and growth on interfaces lead to the formation of three-dimensional heterogeneous structures so-called biofilms. The cells dwelling in these structures are held together by physical interactions mediated by a network of extracellular polymeric substances. Bacterial biofilms impact many human activities and the understanding of their properties is crucial for a better control of their development — maintenance or eradication — depending on their adverse or beneficial outcome. This paper describes a novel methodology aiming to measure in situ the local physical properties of the biofilm that had been, until now, examined only from a macroscopic and homogeneous material perspective. The experiment described here involves introducing magnetic particles into a growing biofilm to seed local probes that can be remotely actuated without disturbing the structural properties of the biofilm. Dedicated magnetic tweezers were developed to exert a defined force on each particle embedded in the biofilm. The setup is mounted on the stage of a microscope to enable the recording of time-lapse images of the particle-pulling period. The particle trajectories are then extracted from the pulling sequence and the local viscoelastic parameters are derived from each particle displacement curve, thereby providing the 3D-spatial distribution of the parameters. Gaining insights into the biofilm mechanical profile is essential from an engineer''s point of view for biofilm control purposes but also from a fundamental perspective to clarify the relationship between the architectural properties and the specific biology of these structures.     

Biofilm structure differentiation based on multi-resolution analysis

Quantitative parameters for describing the morphology of biofilms are crucial towards establishing the influence of growing conditions on biofilm structure. Parameters used in earlier studies generally fail to differentiate complex three-dimensional structures. This article presents a novel approach of defining a parameter vector based on the energy signature of multi-resolution analysis, which was applied to differentiating biofilm structures from confocal laser scanning microscopy (CLSM) biofilm images. The parameter vector distinguished differences in the spatial arrangements of synthetic images. For real CLSM images, this parameter vector detected subtle differences in biofilm structure for three sample cases: (1) two adjacent images of a CLSM stack; (2) two partial stacks from the same CLSM stack with equal numbers of images but spatially offset by one image; and (3) three complete CLSM stacks from different bacterial strains. It was also observed that filtering the noise in CLSM images enhanced the sensitivity of the differentiation using our parameter vector.     

Automated Confocal Laser Scanning Microscopy and Semiautomated Image Processing for Analysis of Biofilms

The purpose of this study was to develop and apply a quantitative optical method suitable for routine measurements of biofilm structures under in situ conditions. A computer program was designed to perform automated investigations of biofilms by using image acquisition and image analysis techniques. To obtain a representative profile of a growing biofilm, a nondestructive procedure was created to study and quantify undisturbed microbial populations within the physical environment of a glass flow cell. Key components of the computer-controlled processing described in this paper are the on-line collection of confocal two-dimensional (2D) cross-sectional images from a preset 3D domain of interest followed by the off-line analysis of these 2D images. With the quantitative extraction of information contained in each image, a three-dimensional reconstruction of the principal biological events can be achieved. The program is convenient to handle and was generated to determine biovolumes and thus facilitate the examination of dynamic processes within biofilms. In the present study, Pseudomonas fluorescens or a green fluorescent protein-expressing Escherichia coli strain, EC12, was inoculated into glass flow cells and the respective monoculture biofilms were analyzed in three dimensions. In this paper we describe a method for the routine measurements of biofilms by using automated image acquisition and semiautomated image analysis.     

Structure of Proteus mirabilis biofilms grown in artificial urine and standard laboratory media

Proteus mirabilis is a urinary pathogen that can differentiate from a swimmer cell into a swarmer cell morphotype and can form biofilms on the surfaces of urinary catheters. These biofilms block these catheters due to crystals trapped within these structures. The effect of encrustation on biofilm formation and structure has not been studied using confocal scanning laser microscopy (CSLM). Therefore, a comparison of biofilm structure in artificial urine (AU) and laboratory media was undertaken. We compared the structure of P. mirabilis biofilms in AU and Luria-Bertani broth using CSLM and 3D imaging. Biofilms grown in Luria-Bertani broth formed mushroom structures at 24 h and contained nutrient channels. AU biofilms were observed to form a different structure at 24 h. AU biofilm structure was observed to be a flat layer, almost devoid of nutrient channels. Swarmer cells were observed protruding out of the biofilm into the bulk fluid. This could be due to nutrient depravation within the biofilm or a means of further colonizing the surface. This study has demonstrated that two markedly different biofilm structures are formed, depending on the growth media utilized.     

Experimental and Computational Investigation of Biofilm Formation by Rhodopseudomonas palustris Growth under Two Metabolic Modes

We examined biofilms formed by the metabolically versatile bacterium Rhodopseudomonas palustris grown via different metabolic modes. R. palustris was grown in flow cell chambers with identical medium conditions either in the presence or absence of light and oxygen. In the absence of oxygen and the presence of light, R. palustris grew and formed biofilms photoheterotrophically, and in the presence of oxygen and the absence of light, R. palustris grew and formed biofilms heterotrophically. We used confocal laser scanning microscopy and image analysis software to quantitatively analyze and compare R. palustris biofilm formation over time in these two metabolic modes. We describe quantifiable differences in structure between the biofilms formed by the bacterium grown heterotrophically and those grown photoheterotrophically. We developed a computational model to explore ways in which biotic and abiotic parameters could drive the observed biofilm architectures, as well as a random-forest machine-learning algorithm based on structural differences that was able to identify growth conditions from the confocal imaging of the biofilms with 87% accuracy. Insight into the structure of phototrophic biofilms and conditions that influence biofilm formation is relevant for understanding the generation of biofilm structures with different properties, and for optimizing applications with phototrophic bacteria growing in the biofilm state.     

Concurrent Quantification of Cellular and Extracellular Components of Biofilms

Confocal laser scanning microscopy (CLSM) is a powerful tool for investigation of biofilms. Very few investigations have successfully quantified concurrent distribution of more than two components within biofilms because: 1) selection of fluorescent dyes having minimal spectral overlap is complicated, and 2) quantification of multiple fluorochromes poses a multifactorial problem. Objectives: Report a methodology to quantify and compare concurrent 3-dimensional distributions of three cellular/extracellular components of biofilms grown on relevant substrates. Methods: The method consists of distinct, interconnected steps involving biofilm growth, staining, CLSM imaging, biofilm structural analysis and visualization, and statistical analysis of structural parameters. Biofilms of Streptococcus mutans (strain UA159) were grown for 48 hr on sterile specimens of Point 4 and TPH³ resin composites. Specimens were subsequently immersed for 60 sec in either Biotène PBF (BIO) or Listerine Total Care (LTO) mouthwashes, or water (control group; n=5/group). Biofilms were stained with fluorochromes for extracellular polymeric substances, proteins and nucleic acids before imaging with CLSM. Biofilm structural parameters calculated using ISA3D image analysis software were biovolume and mean biofilm thickness. Mixed models statistical analyses compared structural parameters between mouthwash and control groups (SAS software; α=0.05). Volocity software permitted visualization of 3D distributions of overlaid biofilm components (fluorochromes). Results: Mouthwash BIO produced biofilm structures that differed significantly from the control (p<0.05) on both resin composites, whereas LTO did not produce differences (p>0.05) on either product. Conclusions: This methodology efficiently and successfully quantified and compared concurrent 3D distributions of three major components within S. mutans biofilms on relevant substrates, thus overcoming two challenges to simultaneous assessment of biofilm components. This method can also be used to determine the efficacy of antibacterial/antifouling agents against multiple biofilm components, as shown using mouthwashes. Furthermore, this method has broad application because it facilitates comparison of 3D structures/architecture of biofilms in a variety of disciplines.     

Homogenization of Pseudomonas aeruginosa PAO1 biofilms visualized by freeze‐substitution electron microscopy

A knowledge of the mechanical properties of bacterial biofilms is required to more fully understand the processes of biofilm formation such as initial adhesion or detachment. The main contribution of this article is to demonstrate the use of homogenization techniques to compute mechanical parameters of Pseudomonas aeruginosa PAO1 biofilms. For this purpose, homogenization techniques are used to analyze freeze substitution electron micrographs of the biofilm cross‐sections. The concept of a representative volume element and the study about his representativeness allows us to determine the optimal size in order to analyze these biofilm images. Results demonstrate significant heterogeneities with respect to stiffness and these can be explained by varying cell density distribution throughout the bacterial biofilms. These stiffness variations lead to different mechanical properties along the height of the biofilm. Moreover, a numerical shear stress test shows the impact of these heterogeneities on the detachment process. Several modes of detachment are highlighted according to the local strain energy in the different parts of the biofilm. Knowing where, and how, a biofilm may detach will allow better prediction of accumulation and biomass detachment. Biotechnol. Bioeng. 2013; 110: 1405–1418. © 2012 Wiley Periodicals, Inc.     

Objective threshold selection procedure (OTS) for segmentation of scanning laser confocal microscope images

The determination of volumes and interface areas from confocal laser scanning microscopy (CLSM) images requires the identification of component objects by segmentation. An automated method for the determination of segmentation thresholds for CLSM imaging of biofilms was developed. The procedure, named objective threshold selection (OTS), is a three-dimensional development of the approach introduced by the popular robust automatic threshold selection (RATS) method. OTS is based on the statistical properties of local gray-values and gradients in the image. By characterizing the dependence between a volumetric feature and the intensity threshold used for image segmentation, the former can be determined with an arbitrary confidence level, with no need for user intervention. The identification of an objective segmentation procedure renders the possibility for the full automation of volume and interfacial area measurement. Images from two distinct biofilm systems, acquired using different experimental techniques and instrumental setups were segmented by OTS to determine biofilm volume and interfacial area. The reliability of measurements for each case was analyzed to identify optimal procedure for image acquisition. The automated OTS method was shown to reproduce values obtained manually by an experienced operator.     

Quantifying biofilm structure: facts and fiction

There is no doubt among biofilm researchers that biofilm structure is important to many biofilm processes, such as the transport of nutrients to deeper layers of the biofilm. However, biofilm structure is an elusive term understood only qualitatively, and as such it cannot be directly correlated with any measurable parameters characterizing biofilm performance. To correlate biofilm structure with the parameters characterizing biofilm performance, such as the rate of nutrient transport within the space occupied by the biofilms, biofilm structure must first be quantified and expressed numerically on an appropriate scale. The task of extracting numerical parameters quantifying biofilm structure relies on using biofilm imaging and image analysis. Although defining parameters characterizing biofilm structure is relatively straightforward, and multiple parameters have been described in the computer science literature, interpreting the results of such analyses is not trivial. Existing computer software developed by several research groups, including ours, for the sole purpose of analyzing biofilm images helps quantify parameters from biofilm images but does nothing to help interpret the results of such analyses. Although computing structural parameters from biofilm images permits correlating biofilm structure with other biofilm processes, the meaning of the results is not obvious. The first step to understanding the quantification of biofilm structure, developing image analysis, methods to quantify information from biofilm images, has been made by several research groups. The next step is to explain the meaning of these analyses. This presentation explains the meaning of several parameters commonly used to characterize biofilm structure. It also reviews the authors' research and experience in quantifying biofilm structure and their attempts to quantitatively relate biofilm structure to fundamental biofilm processes.     

Biofilm structure differentiation based on multi-resolution analysis

Quantitative parameters for describing the morphology of biofilms are crucial towards establishing the influence of growing conditions on biofilm structure. Parameters used in earlier studies generally fail to differentiate complex three-dimensional structures. This article presents a novel approach of defining a parameter vector based on the energy signature of multi-resolution analysis, which was applied to differentiating biofilm structures from confocal laser scanning microscopy (CLSM) biofilm images. The parameter vector distinguished differences in the spatial arrangements of synthetic images. For real CLSM images, this parameter vector detected subtle differences in biofilm structure for three sample cases: (1) two adjacent images of a CLSM stack; (2) two partial stacks from the same CLSM stack with equal numbers of images but spatially offset by one image; and (3) three complete CLSM stacks from different bacterial strains. It was also observed that filtering the noise in CLSM images enhanced the sensitivity of the differentiation using our parameter vector.     

In vivo detection of endotracheal tube biofilms in intubated critical care patients using catheter‐based optical coherence tomography

The formation of biofilms in the endotracheal tubes (ETTs) of intubated patients on mechanical ventilation is associated with a greater risk of ventilator‐associated pneumonia and death. New technologies are needed to detect and monitor ETTs in vivo for the presence of these biofilms. Longitudinal OCT imaging was performed in mechanically ventilated subjects at 24‐hour intervals until extubation to detect the formation and temporal changes of in vivo ETT biofilms. OCT‐derived attenuation coefficient images were used to differentiate between mucus and biofilm. Extubated ETTs were examined with optical and electron microscopy, and all imaging results were correlated with standard‐of‐care clinical test reports. OCT and attenuation coefficient images from four subjects were positive for ETT biofilms and were negative for two subjects. The processed and stained extubated ETTs and clinical reports confirmed the presence/absence of biofilms in all subjects. Our findings confirm that OCT can detect and differentiate between biofilm‐positive and biofilm‐negative groups (P < 10^?5). OCT image‐based features may serve as biomarkers for direct in vivo detection of ETT biofilms and help drive investigation of new management strategies to reduce the incidence of VAP.      

Applying the digital image correlation method to estimate the mechanical properties of bacterial biofilms subjected to a wall shear stress

A digital image correlation (DIC) method was applied to characterize the mechanical behavior of Pseudomonas aeruginosa biofilms in response to wall shear stress using digital video micrographs taken from biofilm flow cells. The appearance of the biofilm in the transmitted light photomicrographs presented a natural texture which was highly conducive to random encoding for DIC. The displacement fields were calculated for two biofilm specimens. The DIC method concurred with previous analysis showing that biofilms exhibit viscoelastic behavior, but had the advantage over simple length measurements of longitudinal strain that it could precisely measure local strains in length (x) and width (y) within biofilm clusters with a 2 μm resolution as a function of time and wall shear stress. It was concluded that DIC was more accurate at measuring elastic moduli than simple length measurements, but that time-lapse 3D images would enable even more accurate estimates to be performed.     

An Aufwuchs Chamber Slide for High-Resolution Confocal Laser Scanning Microscopy and Stereo Imaging of Microbial Communities in Natural Biofilms

Aufwuchs chamber slides were constructed by attaching a silicone rubber gasket to a glass slide with epoxy cement. For biofilm growth, the slides were suspended in Cayuga Lake near Ithaca, NY, for 27 days. Biofilms in the chamber were stained with 0.05% acridine orange. After rinsing, the chamber was filled with molten 1% agarose to stabilize filaments and delicate polymer structures at the biofilm surface. Areas of biofilm ~0.5 mm thick on the inner face of the wall of the chamber were selected for side-on optical sectioning in a confocal laser scanning microscope (CLSM). Stacks of high-resolution optical images captured by the CLSM z-sectioning software, were used to create left-right stereo image pairs. At low magnification the stereo pairs showed 3-D details of the microbial landscape in the mature biofilms. Channels, pores, and other structural features of the biofilm matrix were observed in peripheral regions. Higher magnification images revealed the 3-D distribution of specific biofilm components such as filaments of sheathed bacteria projecting outward into the liquid milieu, and organic coatings, including bacterial cells on the surfaces of mineral particles.     

Measuring local flow velocities and biofilm structure in biofilm systems with magnetic resonance imaging (MRI)

The characterization of substrate transport in the bulk phase and in the biofilm matrix is one of the problems which has to be solved for the verification of biofilm models. Additionally, the surface structure of biofilms has to be described with appropriate parameters. Magnetic resonance imaging (MRI) is one of the promising methods for the investigation of transport phenomena and structure in biofilm systems. The MRI technique allows the noninvasive determination of flow velocities and biofilm structures with a high resolution on the sub-millimeter scale. The presented investigations were carried out for defined heterotrophic biofilms which were cultivated in a tube reactor at a Reynolds number of 2000 and 8000 and a substrate load of 6 and 4 g/m2d glucose. Magnetic resonance imaging provides both structure data of the biofilm surface and flow velocities in the bulk phase and at the bulk/biofilm interface. It is shown that the surface roughness of the biofilms can be determined in one experiment for the complete cross section of the test tubes both under flow and stagnant conditions. Furthermore, the local shear stress was calculated from the measured velocity profiles. In the investigated biofilm systems the local shear stress at the biofilm surface was up to 3 times higher compared to the mean wall shear stress calculated on the base of the mean flow velocity.     

Textural fingerprints: a comprehensive descriptor for biofilm structure development

Automated tools to determine biofilm structure are necessary to interpret large time series of biofilm images. Image analysis based on the evaluation of Spatial Gray Level Dependence Matrices (SGLDM) enabled us to monitor biofilm structure development in response to external disturbances (i.e., periodic increases of wall shear stress) at a large scale (i.e., >1 mm). We applied our method to an experiment conducted in an annular reactor over a 10-week period. Six states of biofilm development were differentiated by their unique structure. Previous exposure to rapidly increased shear influenced the resulting biofilm structure after additional shear increases. In addition, on the scale of the biofilm images, the biofilm structure after a shear increase was spatially heterogeneous and resulted in spatially differentiated regrowth after detachment at different locations in the biofilm. SGLDM was developed further as an alternative to approaches based on image binarization as binarization leads to information loss for low-magnification and low-resolution images. During post-processing of image data, structural states of biofilm development were identified by K-means clustering and data display in Principal Component plots. Quantitatively selected representative images were used to illustrate the meaning of the clusters. Post-treatment of image data was essential for managing several thousands of raw biofilm images and therefore improved the usefulness of the image analysis.