Image analysis with Tissue Recognition for Surface Quantification

The Tissue Recognition for Surface Quantification program enables the quantification of the proportion of an area of tissue in relation to another, in a region of analysis (ROA) of a digital slide. The program is based on the characteristics of the tissues found in the image. In some areas, such as dermatology, the algorithm also enables the average thickness of the analyzed structures to be measured.

The protocol :

The stages of image analysis with the Tissue Recognition for Surface Quantification algorithm are as follows :

  • Creation of a learning model
    The RFT tool enables a learning base to be created, which then enables those tissues presenting the desired characteristics to be identified. The user then stains the different tissues using a brush so that the algorithm can automatically determine the classificatory characteristics.
  • Selection of categories for analysis
    The results of the learning algorithm-based image categorization are shown in order to enable the user to select the categories they wish to analyze and compare.
  • The specific surface explored
    The categories selected constitute the area of interest that is to be compared (Numerator).
  • The surface of reference
    The categories selected represent the surface of comparison (Denominator).
  • Mask post-processing
    The results of the image processing are displayed in accordance with the selected categories. There are tools for carrying out post-processing of the analysis at this stage such as adjustment of object size, contour smoothing, etc.

The specific surface explored compared to the total surface, referred to at the surface of reference, is expressed as a percentage. The results also present the area and the average thickness of the tissues for each of the analyzed surfaces.

Discover how to create an RFT project