Digital Image Analyses

Digital images may be collected by a wide range of options, digital camera or digital video on wide field epifluorescence microscope, 1P-LSM, and 2P-LSM. Collection and analyses of synchrotron images is a specialized area not covered in detail in this chapter.25,26 Once an image series or image stack is generated there are many options to analyse these images and depending on the specific requirements many commercial or freeware systems can be chosen. Key points to consider include: gray level resolution, programming language, capacity for modification, design of macros or plug-ins, and memory requirements, capacity to perform operations on serial image stacks produced in confocal microscopy, or to perform object-based image analysis. Software may be from the microscope company directly or special software companies. All the major companies constantly extend the features on their microscope software. However, in general no software is suitable for every data set and can perform every type of analysis. Consequently, the data has to be treated with different software. The most sophisticated general packages are Imaris (Bitplane), Amira (TGS), and Volocity (Improvision). All of these packages have the usual options plus rendering capacity. Others available include Quanti-met System, Leica (Heidelburg, Germany), MicroVoxel (Indec Systems, Sunnyvale, California), VoxelView (Vital Images, Fairfield, Iowa), or VoxBlast (VayTek, Inc., Fairfield, Iowa), a complete listing of commercial image analyses software may be found most easily by an internet search. Alternately freeware may meet the requirements, the following are widely used systems, Comstat, a program for windows platforms (see http://www.im.dtu.dk/comstat/), NIH Image for Apple platforms (http://rsb.info.nih.gov/nih-image/), or the Windows-based version (ScionImagePC at www.scioncorp.com), and the new Java version ImageJ from NIH or ImageTool (http://ddsdx.uthscsa.edu/dig/itdesc.htlm). There is also Linux based software produced by the French INRA, Nantes, called QUANT3D. An example of how macros developed in NIH Image were used to achieve quantitative measurements in FLBA is provided in Section 6.5.1.

6.5.1 Quantitative In Situ Lectin Analyses

Neu et al.16 provided a detailed analysis of lectin binding in complex systems and proposed a standardized method for digital imaging, image analyses, and calculation of lectin binding. Image analysis was used to define the area of the biofilm binding a specific lectin. In addition, the average gray value of the defined area was determined. These two parameters were used to quantify the area binding a specific lectin according to Equation (6.1).

TA x AGV x 100

255 x 393216

% ICBA is the Intensity Corrected Binding Area, TA is the Thresholded Area of lectin binding, AGV is the Average Gray Value within thresholded area, 255 the gray value of saturated pixels, and 393216 is the number of pixels in a full image (768 x 512).

They noted the importance of incubation time, lectin concentration, the nature of the fluor labeling, presence of carbohydrate inhibition, order of addition, and lectin interactions. An incubation time of 20 min was found to be sufficient; tests indicated that fluorescein isothiocyanate (FITC) conjugated lectins had more specific binding characteristics than tetramethyl rhodamine isothiocyanate (TRITC) or cyanine dye (CY5) labeled lectins. They concluded that the selection of a panel of lectins for investigating the EPS matrix required a full evaluation of their behavior in the micro-bial system to be studied. Neu et al.16 used macros developed in NIH Image and the above equation to analyze 1P-LSM image stacks and determine the quantitative abundance of specific lectin binding sites in river biofilms. Neu et al.18 has applied this approach to examine the impact of nutrients on the glycoconjugate make up of the EPS of river biofilms an approach that may be easily applied to flocs.

0 0

Post a comment