PanNuke Dataset Extension, Insights and Baselines

Abstract

The emerging area of computational pathology (CPath) is ripe ground for the application of deep learning (DL) methods to healthcare due to the sheer volume of raw pixel data in whole-slide images (WSIs) of cancerous tissue slides. However, it is imperative for the DL algorithms relying on nuclei-level details to be able to cope with data from `the clinical wild', which tends to be quite challenging. We study, and extend recently released PanNuke dataset consisting of ~200,000 nuclei categorized into 5 clinically important classes for the challenging tasks of segmenting and classifying nuclei in WSIs. Previous pan-cancer datasets consisted of only up to 9 different tissues and up to 21,000 unlabeled nuclei and just over 24,000 labeled nuclei with segmentation masks. PanNuke consists of 19 different tissue types that have been semi-automatically annotated and quality controlled by clinical pathologists, leading to a dataset with statistics similar to the clinical wild and with minimal selection bias. We study the performance of segmentation and classification models when applied to the proposed dataset and demonstrate the application of models trained on PanNuke to whole-slide images. We provide comprehensive statistics about the dataset and outline recommendations and research directions to address the limitations of existing DL tools when applied to real-world CPath applications.

Work in progress

References in corpus (7)

• Detecting Cancer Metastases on Gigapixel Pathology Images
• Development and Validation of a Deep Learning Algorithm for Improving Gleason Scoring of Prostate Cancer
• Why rankings of biomedical image analysis competitions should be interpreted with care
• Micro-Net: A unified model for segmentation of various objects in microscopy images
• Measuring the tendency of CNNs to Learn Surface Statistical Regularities
• Terabyte-scale Deep Multiple Instance Learning for Classification and Localization in Pathology
• NuClick: From Clicks in the Nuclei to Nuclear Boundaries

Cited by in corpus (17)

• U-Net and its variants for medical image segmentation: theory and applications
• Recent Progress in Transformer-based Medical Image Analysis
• CryoNuSeg: A Dataset for Nuclei Instance Segmentation of Cryosectioned H&E-Stained Histological Images
• NuCLS: A scalable crowdsourcing, deep learning approach and dataset for nucleus classification, localization and segmentation
• BiomedParse: a biomedical foundation model for image parsing of everything everywhere all at once
• CPP-Net: Context-aware Polygon Proposal Network for Nucleus Segmentation
• Recommendations on test datasets for evaluating AI solutions in pathology
• Dataset of Segmented Nuclei in Hematoxylin and Eosin Stained Histopathology Images of 10 Cancer Types
• NuClick: A Deep Learning Framework for Interactive Segmentation of Microscopy Images
• Now You See It, Now You Dont: Adversarial Vulnerabilities in Computational Pathology
• Unlocking the Potential of Digital Pathology: Novel Baselines for Compression
• Nuclei Grading of Clear Cell Renal Cell Carcinoma in Histopathological Image by Composite High-Resolution Network
• Impact of imperfect annotations on CNN training and performance for instance segmentation and classification in digital pathology
• CoNIC: Colon Nuclei Identification and Counting Challenge 2022
• PointNu-Net: Keypoint-assisted Convolutional Neural Network for Simultaneous Multi-tissue Histology Nuclei Segmentation and Classification
• Unified Medical Image Segmentation with State Space Modeling Snake
• Learning Interpretable Microscopic Features of Tumor by Multi-task Adversarial CNNs To Improve Generalization