Image analysis of particle dispersions in microscopy images of cryo‐sectioned sausages

Abstract

Summary:Two feature extraction methods, the threedimensional (3-D) local box-counting method and the area distribution method, are presented to describe the fat dispersion pattern on digital microscopy images of cryo-sectioned sausages. Both methods calculate whole arrays of variables for each microscopy image. The 3-D box-counting method calculates scale dependent (local) dimensions. This is in contrast to common fractal methods, which are univariate. Principal component analysis (PCA) was used to show that different sausages yield different fat dispersion patterns. Partial least square regression (PLS) shows that there is a correlation between the variables gained with both methods and the fat content.

Identification of fat, protein matrix, and water/starch on microscopy images of sausages by a principal component analysis‐based segmentation scheme

“…The production of the sausages runs along similar lines as the procedure described in Kohler et al (2001). The raw materials were frozen and thawed pork meat, pork fat, NaCl (1.8%), starch (5%), and ice water, while the target meat protein and fat levels were 10 and 15%, respectively.…”

“…The batter temperature was 15°C and thermal processing ended at a 78°C internal endpoint temperature of the sausages. Blocks of sausages <0.5 cm 3 were frozen in liquid nitrogen and sectioned, 10µm thick, in a Leitz cryostat (Wild Leitz GmbH, Heerbrugg, Switzerland) at -22°C to -25°C and placed on microscopic slides (Kohler et al 2001, Ofstad et al 1995. Subsequently, the sections were covered with coverslips, 10 µl Nile red (9-diethylamino-5H-benzophen-oxazin-5-one) dissolved in acetone (0.1 mg/ml) was settled, and the aceton was evaporated before mounting with glycerol (70µl).…”

“…Traditionally, analyses of images have been based on counting objects, on measuring object areas, and by relating the gained information to chemical data by the use of univariate techniques (Carballo et al 1996, Koolmees et al 1989, Lee 1985. Recent papers (Egelandsdal et al 1999, Gao and Tan 1996, Kohler et al 2001, Kvaal et al 1998) have evaluated several multivariate methods for extracting features with respect to their ability to extract information about relevant parameters in food systems. The conclusions were that no single method was superior to the others with respect to predicting sensory attributes; that is, the method depends on the application.…”

“…The conclusions were that no single method was superior to the others with respect to predicting sensory attributes; that is, the method depends on the application. Kohler et al (2001) investigated the relationship between fat content and chopping level and the microstructure on the microscopy images of cryosectioned sausages by two different automated feature extraction methods; the three-dimensional local box-counting method and the area distribution method. Principal component analyses (PCA) and partial least-squares regressions (PLS) were used to show that sausages containing 15, 20, and 25% fat yield different fat dispersion patterns.…”

FTIR Imaging for Structural Analysis of Frankfurter Sausages Subjected to Salt Reduction and Salt Substitution

“…10 However, microstructural examination is tedious and involves staining of individual components on separate sections. 10,11 As a powerful and versatile analytical technique, FTIR microspectroscopy enables monitoring structural changes in different components simultaneously, such as bulk proteins, fats, and carbohydrates. This is done by acquiring single FTIR spectra of tissue components such as myofibers in microscopic mode.…”