• 2022-06
  • 2022-05
  • 2022-04
  • 2021-03
  • 2020-08
  • 2020-07
  • 2018-07
  • br Preparation of database of cell


    3.2. Preparation of database of cell patterns
    The typical microscopic field of view displaying a tissue frag-ment usually contains 50 to 100 cells, represented by their nuclei. The view field size used in the experiments was 2070 × 1548 pix-els. An experienced expert from the pathology department manu-ally marked the boundaries of the complete cells, which were then included in the cell pattern database. The database was automat-ically saved for use in the further stage of the experiments. Each cell pattern image was saved as a separate unit. Fig. 3 depicts such a situation for the field of view (left image), and the corresponding cell pattern database is displayed on the right. 
    Fig. 3. Creation of cell pattern database: (a) analyzed field of view and (b) database of individual segmented cell patterns.
    The size of the individual pattern cell image was then cut to a square with the size
    where α and β are the cell width and height, respectively. A small surrounding area of 30 pixels on each side was included. More-over, the small area of pixels representing the gene biomarkers (HER2: red and CEN17: green) in the cell patterns were detected using a fuzzy pattern recognition algorithm, as presented in Les, Markiewicz, Osowski, Kozlowski, and Jesiotr (2016). The pixels in these areas were filtered using a median filter, and the new val-ues for these pixels were assigned (Gonzalez & Woods, 2011). The yellow-green stroma pixels were also removed from the image us-ing the “map of colors” method (Gonzalez & Woods, 2011). Only the Tacrolimus cleaned from the stroma that were not deformed were used as patterns in the final cell reconstruction system. The binary masks, representing the boundaries of the pattern cells, were also stored.
    3.3. Pre-segmentation of test cells
    A test cell was defined as a cell for which we wished to check the possible reconstruction of its expected shape. Moreover, we wanted to assess the quality of the reconstruction and parameters characterizing the overlapped area of the deformed cells.
    The first pre-processing work included preliminary segmenta-tion of the cells. The aim was to identify the approximate bound-ary of all cells in the analyzed field of view. This could be achieved using any method belonging to traditional image thresholding techniques; for example, an algorithm based on watershed trans-formation (Xing & Yang, 2016). The result of this operation was the mask, representing the positions and edges of all detected cells. A natural side effect of this transformation was deformation of the cell shapes, resulting from the nonlinear distribution of the pixel brightness or the overlapping problem. Cell reconstruction would allow restoring the original shape of the cells. The final step of the test cell preparation was similar to that of the cell patterns, as de-scribed in the previous step. In this stage, the image of each test cell was cut to a size of ρ = max (α, β ) + 30 pixels and stored as a separate unit. The set of test cells represented the candidates for reconstruction.
    3.4. Similarity and sensitivity measures for two images
    The measures of the sensitivity and similarity of patterns play important roles in speeding up the final reconstruction procedure.
    The similarity of two RGB images is assessed based on the distance between these images. The Euclidean measure of distance between the pixel brightness values of two RGB images, with T representing the tested cell and P representing the pattern cell, was applied. This measure is defined as follows:
    where T(i) and P(i) represent the brightness values of the ith pixel in the test and pattern images, respectively, while the indices R, G, and B denote the red, green, and blue channels, respectively. Based on this distance, the similarity measure of the test and pattern im-ages is defined as follows: