Automated identification of stained cells in tissue sections using digital image analysis.

OBJECTIVE: To develop a novel automated image analysis system to differentiate immunohistochemically stained cells from background. STUDY DESIGN: Cell segmentation was performed by applying global thresholding algorithms to find an approximate threshold at which cells could be separated from background followed by a novel refinement algorithm to erode edge pixels of the region. To separate overlapping cells, a new decomposition method was developed that uses both semantic knowledge and high-level relational information. Both the cell segmentation and separation methods were evaluated on images of stained tissue sections and the manually outlined cell areas and numbers compared to the computed. RESULTS: Macrophage areas computed at the first stage by Otsu's algorithm did not differ significantly (P = .07) from those traced manually, while the areas computed by Kittler's and Kurita's algorithms did not agree (P < .01). Both Otsu's and Kurita's algorithms performed well when combined with edge pixel erosion. Kittler's algorithm proved unsuccessful even with edge erosion. Comparison of the computed and manually determined cell numbers showed a significant correlation, and regression analysis resulted in the unity curve. CONCLUSION: A combination of global thresholding and a novel edge erosion technique allowed identification of immunohistochemically stained macrophages; the computed cell areas agreed with the manual results.

Three-dimensional imaging of vasculature and parenchyma in intact rodent organs with X-ray micro-CT.

A microcomputed tomography (micro-CT) scanner, which generates three-dimensional (3-D) images consisting of up to a billion cubic voxels, each 5-25 micron on a side, and which has isotropic spatial resolution, is described. Its main components are a spectroscopic X-ray source that produces selectable primary emission peaks at approximately 9, 18, or 25 keV and a fluorescing thin crystal plate that is imaged (at selectable magnification) with a lens onto a 2.5 x 2.5-cm, 1,024 x 1,024-pixel, charge-coupled device (CCD) detector array. The specimen is positioned close to the crystal and is rotated in 721 equiangular steps around 360 degrees between each X-ray exposure and its CCD recording. Tomographic reconstruction algorithms, applied to these recorded images, are used to generate 3-D images of the specimen. The system is used to scan isolated, intact, fixed rodent organs (e.g., heart or kidney) with the image contrast of vessel lumens enhanced with contrast medium. 3-D image display and analysis are used to address physiological questions about the internal structure-to-function relationships of the organs.

Quantitative analysis of inflammatory cells in aortic atherosclerosis of young adults.

Cellular analysis of aortic atherosclerotic lesions has been pursued extensively in recent years, although most of these investigations have involved the detection of inflammatory cells in chronically diseased tissue or artificially induced atherosclerosis in an animal model. Few studies have attempted to quantify accurately, using computer analysis systems, the degree of cellular infiltration in a statistically significant number of samples, in tissue from young adults. In this study, segments of human aortae were collected at autopsy from 29 individuals ranging in age from 15 to 35 years. The tissue was embedded in paraffin and stained using routine histological and automated immunohistochemical staining techniques. The sections were evaluated using advanced image analysis techniques to investigate the differences in cellular composition and cell activation between the dorsal and ventral aspects of the human aorta and to correlate these findings to the age of the subjects. These regions have been previously shown to have a high (dorsal) and low (ventral) probability of developing sudanophilic lesions. Our data demonstrated that statistically different cell populations exist in the dorsal and ventral regions of each vessel. The dorsal aspect (i.e., high-probability region) had a greater number of HAM56(+) (36.9% increase,p = 0.0002) and HLA-DRα(+) cells (44.2% increase,p = 0.0035) than did the ventral surface (i.e., low-probability region), although there were no significant differences in the number of CD43(+) lymphocytes. When grouped according to age, results showed significant increases in the dorsal region when considering HAM56(+) and HLA-DRα(+) cells (p = 0033 and 0.046, respectively). Morphologically, a greater number of foam cell aggregates were found to occur in the dorsal region of the vessel than in the ventral portion. Our results indicate that the microarchitecture and cellular composition of the dorsal and ventral aorta are significantly different, with these variations becoming more marked with age.