This algorithm quantifies in situ hybridization slides. With CISH/FISH, the algorithm permits to segment the nuclei and to measure two probes on a cell by cell basis. The results show a classification of the nuclei based on signal amplification. With RNA scope, the algorithm quantifies the number of dots in each nucleus and sorts the cells according to it.
The stages of image analysis with ISH algorithm are as follows:
- Nuclei segmentation
A nuclei segmentation is performed on luminance on one of the color plans to eliminate the rest of the tissue. A separation algorithm is then applied to individualize the nuclei.
- Genes segmentation
Each gene is segmented on its own color channel. A dot and / or surface approach is available.
For each nucleus, a number of copies of the gene is estimated, eventually including its control markers. Each nucleus is then sorted according to the number of copies or to the amplification ratio.
The results present the number of valid segmented nuclei as well as a class according to the number of copies or to the amplification ratio. A color representation and a gallery mode allow fast validation.