Computerized image analysis

Image analysis can be carried out by means of a plethora of software packages, some of which are supplied by the main optical or electronic microscopies manufacturers (Jeol, Leica, Karl Zeiss, Nikkon, etc.) as a complement to their devices. 

Of course a correct selection of threshold gray level is fundamental to performing a correct analysis of accurate assigned pores. Customarily the gray spectrum is analyzed and the threshold placed centered in the peak to peak valley of the almost bimodal distributions obtained. Unfortunately sometimes the spectra are so flat that this technique is only of relative help to make a correct threshold election [40]. In any case, eye inspection facUitates the process of selection of several reasonable threshold candidates whose outcomes are conveniently averaged. 

---

Physical testing appHcations and methods for fibrous materials are reviewed in the Hterature (101—103) and are generally appHcable to polyester fibers. Microscopic analyses by optical or scanning electron microscopy are useful for evaluating fiber parameters including size, shape, uniformity, and surface characteristics. Computerized image analysis is often used to quantify and evaluate these parameters for quaUty control. 

The conformation change in the power stroke has been studied in two ways (1) cryoelectron microscopy together with computerized image analysis... 

Morphometric and stereological research methods are valuable quantitative tools for toxicological studies. However, these methods can be very labour intensive and therefore are infrequently used for toxicity evaluation studies. Computerized image analysis may solve some of these problems in the fiiture. Other automated quantitative methods, such as flow cytometry, have also been used (Kangasniemi et al., 1990 Toppari et al., 1990). However, these are not routine techniques and are not incorporated in any testing guidelines at present. 

A related approach was used by Ranefall et al. (1998) to quantify images of immunohistochemically stained cell nuclear Ki-67 antigen and cyclin A protein in bladder carcinoma tissue. They combined automatic, computerized image analysis with appropriate controls and reference material. This approach is superior to the automatic method without... 

James, J. D., and Hauer-Jensen, M. 1999. Effects of fixative and fixation time for quantitative computerized image analysis of immunohistochemical staining. J. Histotechnol. 22 109-111. 

Press, M. F., Pike, M. C., Chazin, V. R., Hung, G., Udove, J. A., Marcowicz, M., Danyluk, J., Godolphin, W., Sliwkowski, M., and Akita, R. 1993. Her-2/neu expression in node-negative breast cancer Direct tissue quantitation computerized image analysis and association of overexpression with increased recurrent disease. Cancer Res. 53 4960-4970. 

Figure 2.6 SEM micrograph of the top layer surface of an UF membrane from PAN and pore-size distribution from computerized image analysis (reprinted from [9], with permission from Wiley-VCH, 2006).

A combination of electron microscopy, computerized image analysis, and mechanical testing provides a powerful tool for studying morphology-property relationships in toughened plastics. 

A statistical framework has been established for describing the spatial dispersion of second phase particles in a continuous matrix. Based on this scheme a computerized image analysis method has been developed for characterizing the morphology of toughened plastics. 

Comad, T., Chandler, D., Corless, J., Klintworth, G. (1994). In vivo measurement of comeal angiogenesis with video data acquisition and computerized image analysis. Lab. Invest. 70 426-34. 

After being removed from the centrifuge, samples were examined and analyzed by optical microscopy, scanning electron microscopy (SEM), x-ray mapping, computerized image analysis, and x-ray diffraction analysis (XRD). 

A graded distribution of the ceramic particles for the case of x = 7 was also obtained. However, a major difference between the two cases relates to the size of the alumina particles. The difference between the two systems can be better demonstrated by plots of AljO, particle size distributions for the two cases. These are seen in Figures 3(a) and (b) for x = 6 and 7, respectively. The plotted results were obtained by a computerized image analysis where the average particle size was determined in layers sectioned successively within the FGM zone. In both cases, the particle size decreased by a factor of two from the ceramic side to the metallic side in the FGM zone. However, the AljO, particles in the x = 6 system are significantly larger and... 

Nigar E, Dervan PA. Quantitative assessment of basement membranes in soft tissue tumors Computerized image analysis of laminin and type IV collagen./Pathol. 1998 185 184-187. 

P. R. C. Gascoyne, Y. Huang, R. Pethig, J. Vykoukal and F. F. Becker, Dielectrophoretic separation of mammalian cells studied by computerized image analysis, Meas. Sci. TechnoL, 3, 439-445 (1992). 

High resolution two-dimensional electrophoresis with computerized image analysis and data reduction is the highest resolution method currently available for the analysis of complex protein mixture. In this discussion we review the potential of the method for the analysis of agricultural food products including milk, meat products, and wheat, and present one representative study on the analysis of Newton wheat for electrophoretic variants. 

The types of samples which may be analyzed by 2D6EL are varied. Including virtually all animal tissues, fluids, and cells, and many microbial and plant specimens. The resulting patterns for most of these specimens are complex, and often contain more information than the researcher can immediately utilize, especially since 2D6EL is most often done to study a small subset of the total protein set present. This type of local" analysis may be done visually, or with the aid of a small densitometer (IQ). However, much more complex analyses are possible ("global" analyses) with the aid of a computerized Image analysis system (ID- This is discussed in more detail below. 

If the orientation can be examined by optical microscopy, it can be quantified using computerized image analysis techniques. Anisotropy can be studied by applying an ultrasonic, electrical or radiation signal, which penetrates into the test material in different directions. Gross fibre waviness can be checked by visual inspection. 

At s.d.e. die determination of the time dependence of the t.d.c. is sufficient for the investigation of coalescence. In Refs 27 and 28 this was accomplished through direct visual observation. By using video-enhanced microscopy and computerized image analysis the determination of t.d.c. can be automated. Such automated determination of total droplet number in a dilute DCD-in-water emulsion at the s.d.e. can be recommended as a standard method for the characterization of the elementary act of coalescence. 

Fiber crimp (waviness) affects the ease with which fibers may be spun into yarn, as well as many of the comfort characteristics of cloth made from those yarns. Computerized image analysis techniques can be used to measure the degree of crimp of a fiber. 

Interpretation of the results is again a problem. Tests using home made monoclonal antibodies are suspect until the antisera is made available to other investigators for conformation. The use of well characterized antisera from companies which supply to others is better at this stage. Since the tissue section is examined and scored by an observer, the data from these studies are not really available for analysis by the reader. Computerized image analysis techniques are still not widely used. Thus, in evaluating the results, possible bias of the investigator must be taken into account. 

Fractographic analyses of CF must be improved, including quantitative measurement of crack surface crystallography [119], and computerized image analysis methods to characterize and reconstruct the CF process [120]. 

Every image obtained by any method on a micro- or macro-scale may be subjected to computerized image analysis providing quantitative results, if particular elements in that image can be distinguished from the others. Several systems are known and commercialised by specialized companies, which allows different features of selected elements in an image to be recorded and calculated - area and perimeter, distribution, shape, etc. The possibilities of such a system depend not only upon the technical performance, bnt mainly on available software which allows more or less advanced quantitative analysis (cf. also Section 6.6.2). 

---