Nuclei Instance Segmentation and Classification in Histopathology Images with Stardist

Instance segmentation and classification of nuclei is an important task in\ncomputational pathology. We show that StarDist, a deep learning nuclei\nsegmentation method originally developed for fluorescence microscopy, can be\nextended and successfully applied to histopathology images. This is\nsubstantiated by conducting experiments on the Lizard dataset, and through\nentering the Colon Nuclei Identification and Counting (CoNIC) challenge 2022,\nwhere our approach achieved the first spot on the leaderboard for the\nsegmentation and classification task for both the preliminary and final test\nphase.\n