Recognition and segmentation of tissue classes
The algorithm maps tissues by identifying anatomical structures and regions. It allows to define the regions of analysis to be quantified and the invasion margin of the tumor.
Tissue Recognition analyzes tissues (e.g. stroma vs tumor, glands vs tissue) by machine learning from learning images. The analysis is based on criteria of colors, contours, shapes and textures.
The protocol :
The stages of image analysis with the Tissue Recognition for ROA algorithm are as follows :
- Creation of a learning model
The RFT tool enables a learning base to be created that enables cells presenting certain characteristics to be identified. The user then stains the different objects using a brush so that the algorithm can automatically determine the classificatory characteristics.
- Selection of the classes to be analyzed to make a ROA or to measure surfaces
The results of the learning algorithm-based image categorization are shown. This enables the user to select the categories they wish to analyze. As such, only the area or areas selected will be kept.
The mask created by the algorithm is displayed on the region of analysis (ROA). This latter can then be processed by an image analysis and quantification program using our CaloPix software.