Fractal and Image Analysis of the Microvasculature in Normal Intestinal Submucosa and Intestinal Polyps in Apc^Min/+ Mice

Abstract

Tumors are supported by the development of a unique vascular bed. We used fractal dimension (Db) and image analysis to quantify differences in the complexity of the vasculature in normal intestinal submucosa and intestinal polyps. Apc(Min/+) mice and wild-type mice were perfused with a curable latex compound, intestines sectioned, and images collected via confocal microscopy. The images were analyzed and area (A), perimeter (P), and integrated optical density (IOD) of the normal and tumor vascular beds were measured. The Db, a quantitative descriptor of morphological complexity, was significantly greater for the polyp vasculature from Apc(Min/+) mice than controls. This indicates that the polyp microvasculature is more chaotic than that of the controls, while the IOD and average vascular density values displayed no differences. This suggests the mass of blood volume is equivalent in normal and polyp microvasculature. The lower vascular area-perimeter ratios expressed by the polyp microvasculature suggest it is composed of smaller, more tortuous vessels. These data demonstrate that fractal analysis is applicable for providing a quantitative description of vascular complexity associated with angiogenesis occurring in normal or diseased tissue. Application of Db, IOD, and average density provides a clearer quantification of the complex morphology associated with tissue microvasculature.

Total Abdominal 18F-FDG Uptake Reflects Intestinal Adenoma Burden in Apc Mutant Mice

“…The finding that the ex vivo intestinal 18 F-FDG uptake increased at age 12 wk, compared with at age 9 wk, whereas the number of (both total and large) adenomas did not, strengthens the hypothesis that more advanced dysplastic lesions were well detected by the small-animal PET scanner at age 12 wk. Possibly, adenoma numbers did not increase after age 9-10 wk because angiogenesis could not keep up with the fast-growing adenomas as a result of the chaotic microvasculature observed in 12-wkold Apc Min mice (26), leading to insufficient blood supply and stabilization or even regression of adenomas.…”

Nitric Oxide Modulates Postnatal Bone Marrow-Derived Mesenchymal Stem Cell Migration

“…The box counting method (Fernandez and Jelinek, 2001; Grizzi et al, 2005) was used to determine the fractal dimension (D) of the MSCs monolayer. Of the numerous methods for applying fractal analysis to biological and non-biological systems, the box-counting method was most widely used, and provided a general model for determining ( D ) for diverse systems (Fuseler et al, 2007, 2010; Moledina et al, 2011; Wedman et al, 2015). The box-counting method consisted of a grid of boxes of size e superimposed over the image or the region of interest (ROI) of the image or structure, here individual MSCs, and the number of boxes containing any part of the ROI or structure recorded as N ( e ).…”

“…The box fractal dimension ( D ) could be determined from the slope of the regression line, i.e., ( D ) = Log [ N ( e )]/ Log (1/ e ). The values of D were determined using HarFA mathematical analytical software (Nezadal et al, 2001; Fuseler et al, 2010). The HarFA software for the 10x images assigned mesh sizes of boxes with e values ranging from 2 to 215 pixels and 10 steps within this range were calculated to generate the Log [ N ( e )] vs. Log (1/ e ) lines to determine D. In these 10x images, one pixel was = 1.7786 μm,” (Fuseler and Valarmathi, 2012).…”

“…Using the software's optical calipers, the measurements were refined by setting specific boundary conditions for area and IOD for acceptance of the signal from the individual MSC actin cytoskeleton and to minimize or eliminate the contributions of any non-specific and background staining. This concept of IOD representing the mass or total amount of stained material in an ROI of an image was well-established, and has been previously applied to diverse topics including the translocation of nuclear factor-κB in the nucleus (Fuseler et al, 2006; Rogers and Fuseler, 2007), changes in actin cytoskeleton in cardiac fibroblasts and adult stromal stem cells (Fuseler et al, 2007; Fuseler and Valarmathi, 2012), as well as comparison of the architecture of normal and tumor microvasculature (Fuseler et al, 2010). The IOD measured for the actin cytoskeleton was a representation of the mass and a measurement of the total amount of material in the delineated region.…”

Effects of Noninhibitory Serpin Maspin on the Actin Cytoskeleton: A Quantitative Image Modeling Approach

“…But under-and over-segmentation is a common problem of watershed segmentation, where a strong post-processing is needed to avoid this problem. Recently, several quantitative techniques were developed to quantify the subcellular structures like actin cytoskeleton, fibres like fibronectin, collagen; fractal based image analysis of actin cytoskeleton of neonatal cardiac fibroblast (Fuseler et al 2007) and microvasculature in normal intestinal submucosa (Fuseler et al 2010), quantification of immunostaining using digital image subtraction, blue filter and enhancement (DISBE) method (Bernardo et al 2009) and automated podosome identification and characterization using segmentation and quantitative image modelling (Meddens et al 2013). But to date, a complete segmentation technique is warranted for segmenting cellular cytoskeletal regions to identify variations in subcellular organisations from co-stained nuclei and cytoplasm image samples.…”