• 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • 2021-03
  • br Fig presents an exemplary image


    Fig. 2 presents an exemplary image of the enlarged nucleus, along with the stroma overlapping certain parts of the cell. The original registered image is presented on the left. The middle one illustrates the cell nucleus with an overlapping stroma area, marked in orange. The right-hand image is the cell with the cor-rected shape, with the proper boundary marked by a medical ex-pert.
    A further problem is to evaluate the cell deformation degree. This assessment allows for undertaking a decision as to whether or not the cell should be included in the overall pathological exami-nation. However, no precise guidelines exist specifying the percent-age of the cell that must be visible to treat it as a separate one. Ex-pert pathologists must therefore decide on this individually. More-over, at times, the cell is overlapped by stroma, but the biomarkers shine through a thin layer of stroma. Such a cell should evidently
    Fig. 1. Example of a field of view with intercellular space visible under a micro-scope with 100× lens magnification and 10× internal magnification.
    Fig. 2. Example of cell nucleus with stroma overlapping. Original registered image is presented on left. Middle one illustrates cell nucleus with overlapping stroma area, marked in orange. Right-hand image is cell with corrected shape, with the proper boundary marked by a medical expert.
    be rejected. The proposed method enables such situations to be recognized.
    The presented approach to cell reconstruction is based on the comparison of patterns. In the computer analysis process, the boundary of each cell is compared to specially prepared patterns. Such a comparison allows for calculating the similarity degree of the test cell boundary to these patterns. If both shapes meet spe-cific conditions, defined in the method, it is possible to reconstruct the deformed cell contour using a pattern shape.
    3. Reconstruction method
    The reconstruction of Gly Pro pNA in the entire microscopic field of view is a complex process. In this work, it was divided into two main stages: (1) preparation of the cell pattern database and (2) reconstruction of the tested cells by comparison with the patterns. The aim of the first step was to select the set of effectively seg-mented cells, which would be used as the typical patterns, based on which the tested (often distorted) cells would be converted into the proper shape. This first step was performed only once. The data prepared therein would be used in the second stage for re-constructing the deformed candidate cells, which were identified in the field of view of the analyzed image. The second stage was divided into the following phases:
    • pre-segmentation of the candidate cells,
    • assessment of the similarity and sensitivity measures, and
    • final reconstruction of the cells.
    Each candidate test cell was compared to all pattern cells and the similarity degree was calculated, followed by the reconstruc-tion procedure. The detected pattern cell contour with the great-est similarity was used to define the final shape of the candidate cell. Following possible rotation and scaling, the pattern cell con-tour determined the new reconstructed boundary of the tested cell. The obtained results also allowed for determining the difference between the deformed candidate and reconstructed contours with respect to the size and shape of the overlapped area.
    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 cells 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.