Nuclei Instance Segmentation and Classification in Histopathology Images with Stardist
Martin Weigert, Uwe Schmidt
Abstract
Instance segmentation and classification of nuclei is an impor-tant task in computational pathology. We show that StarDist, a deep learning nuclei segmentation method originally devel-oped for fluorescence microscopy, can be extended and suc-cessfully applied to histopathology images. This is substan-tiated by conducting experiments on the Lizard dataset, and through entering the Colon Nuclei Identification and Counting (CoNIC) challenge 2022, where our approach achieved the first spot on the leaderboard for the segmentation and clas-sification task for both the preliminary and final test phase.
Topics & Concepts
HistopathologyArtificial intelligenceSegmentationComputer scienceImage segmentationPattern recognition (psychology)Computer visionPathologyMedicineAI in cancer detectionDigital Imaging for Blood DiseasesRadiomics and Machine Learning in Medical Imaging