TissueGnostics’ new Deep Neural Network (DNN) was developed to handle nuclear segmentation in extremely dense tissue microenvironments with minimal user input. The shown example was challenging for nuclear segmentation due to the high cellular density, non-homogenous chromatin distribution pattern, and the weak DAPI (blue) signal intensity. The user was able to obtain a precise nuclear segmentation despite these three major issues using the DNN in StrataQuest 7.
The following example was analyzed using a combination of TG’s powerful machine learning-based classifier for a precise detection of morphological entities and the DNN for a more accurate nuclei segmentation. The image is showing a kidney biopsy stained for various markers including DAPI (blue), immune cell marker (white), glomeruli detection marker (red) and a marker to detect the tubules (green).