Gary Harkin

Quantifying biofilm structure using image analysis

We have developed and implemented methods of extracting morphological features from images of biofilms in order to quantify the characteristics of the inherent heterogeneity. This is a first step towards quantifying the relationship between biofilm heterogeneity and the underlying processes, such as mass-transport dynamics, substrate concentrations, and species variations. We have examined two categories of features, areal, which quantify the relative magnitude of the heterogeneity and textural, which quantify the microscale structure of the heterogeneous elements. The feature set is not exhaustive and has been restricted to two-dimensional images to this point. Included in this paper are the methods used to extract the structural information and the algorithms used to quantify the data. The features discussed are porosity, fractal dimension, diffusional length, angular second moment, inverse difference moment and textural entropy. We have found that some features are better predictors of biofilm behavior than others and we discuss possible future directions for research in this area.

Evaluation of biofilm image thresholding methods.

To evaluate biomass distribution in heterogeneous biofilms from their microscope images, it is often necessary to perform image thresholding by converting the gray-scale images to binary images consisting of a foreground of biomass material and a background of interstitial space. The selection of the gray-scale intensity used for thresholding is arbitrary but under the control of the operator, which may produce unacceptable levels of variability among operators. The quality of numerical information extracted from the images is diminished by such variability, and it is desirable to find a method that improves the reproducibility of thresholding operations. Automatic methods of thresholding provide this reproducibility, but often at the expense of accuracy, as they consistently set thresholds that differ significantly from what human operators would choose. The performance of five automatic image thresholding algorithms was tested in this study: (1) local entropy; (2) joint entropy; (3) relative entropy; (4) Renyis entropy; and (5) iterative selection. Only the iterative selection method was satisfactory in that it was consistently setting the threshold level near that set manually. The extraction of feature information from biofilm images benefits from automatic thresholding and can be extended to other fields, such as medical imaging.

Quantifying biofilm structure

This article defines some quantitative parameters for describing the structure of a biofilm. The parameters can be calculated from a two-dimensional cross-sectional image on a plane parallel to the substratum within an in situ biofilm. Such images can be acquired using a confocal scanning laser microscope (CSLM). The parameters will eventually be used for eliciting relationships between the biofilms structure and its biochemical function, and for computer model evaluation. The results shown here indicate that the structural parameters appear to be reaching steady-state conditions as the biofilm grows to a steady state.

Chemical effects of biofilm colonization on 304 stainless steel

Changes in the surface concentrations of the main alloying elements of an as‐received 304 stainless steel, exposed to a mixed culture of biofilm‐forming bacteria under flowing conditions, were observed using Auger electron spectroscopy. In the oxide film close to the bulk stainless steel, there was an enrichment in the relative concentration of Cr with a corresponding decrease in the relative Fe concentration as compared to a control coupon exposed only to sterile media. There were no changes observed in the relative Ni concentration.

Parameterized algorithm decomposition and performance analysis

A decomposition model using two-task replacement sets is used to model algorithm decomposition and algorithm performance. This technique provides a parametric representation of decomposition and performance analysis that is more general and powerful than previous methods. The model is used to investigate the performance of algorithms for MIMD architectures, and the results include statistical descriptions of task and synchronization penalty behavior under decomposition, and analytical representations of algorithm performance. The analysis provides insight into the effect of decomposition on performance at both the local task and the global algorithm levels.<<ETX>>