Detecting Cancer Metastases on Gigapixel Pathology Images

Abstract

Each year, the treatment decisions for more than 230,000 breast cancer patients in the U.S. hinge on whether the cancer has metastasized away from the breast. Metastasis detection is currently performed by pathologists reviewing large expanses of biological tissues. This process is labor intensive and error-prone. We present a framework to automatically detect and localize tumors as small as 100 x 100 pixels in gigapixel microscopy images sized 100,000 x 100,000 pixels. Our method leverages a convolutional neural network (CNN) architecture and obtains state-of-the-art results on the Camelyon16 dataset in the challenging lesion-level tumor detection task. At 8 false positives per image, we detect 92.4% of the tumors, relative to 82.7% by the previous best automated approach. For comparison, a human pathologist attempting exhaustive search achieved 73.2% sensitivity. We achieve image-level AUC scores above 97% on both the Camelyon16 test set and an independent set of 110 slides. In addition, we discover that two slides in the Camelyon16 training set were erroneously labeled normal. Our approach could considerably reduce false negative rates in metastasis detection.

arXiv commentFig 1: normal and tumor patches were accidentally reversed - now fixed. Minor grammatical corrections in appendix, section "Image Color Normalization"

References in corpus (1)

• Deep Learning for Identifying Metastatic Breast Cancer

Cited by in corpus (91)

• A Survey on Deep Learning in Medical Image Analysis
• A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAI
• Machine learning methods for histopathological image analysis
• Deep neural network models for computational histopathology: A survey
• Deep learning-based virtual histology staining using auto-fluorescence of label-free tissue
• Transfusion: Understanding Transfer Learning for Medical Imaging
• Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology
• Grader variability and the importance of reference standards for evaluating machine learning models for diabetic retinopathy
• Whole Slide Images based Cancer Survival Prediction using Attention Guided Deep Multiple Instance Learning Networks
• Development and Validation of a Deep Learning Algorithm for Improving Gleason Scoring of Prostate Cancer
• PhaseStain: Digital staining of label-free quantitative phase microscopy images using deep learning
• Microscope 2.0: An Augmented Reality Microscope with Real-time Artificial Intelligence Integration
• Scalable Private Learning with PATE
• A Comprehensive Review for Breast Histopathology Image Analysis Using Classical and Deep Neural Networks
• RMDL: Recalibrated multi-instance deep learning for whole slide gastric image classification
• PanNuke Dataset Extension, Insights and Baselines
• Artificial Intelligence in Clinical Health Care Applications: Viewpoint
• Weakly Supervised Deep Learning for Thoracic Disease Classification and Localization on Chest X-rays
• Automated sub-cortical brain structure segmentation combining spatial and deep convolutional features
• Multi-scale fully convolutional neural networks for histopathology image segmentation: from nuclear aberrations to the global tissue architecture
• Deep Multi-Magnification Networks for Multi-Class Breast Cancer Image Segmentation
• Whole-Slide Image Focus Quality: Automatic Assessment and Impact on AI Cancer Detection
• Classification of crystallization outcomes using deep convolutional neural networks
• Cancer Metastasis Detection With Neural Conditional Random Field
• MCUa: Multi-level Context and Uncertainty aware Dynamic Deep Ensemble for Breast Cancer Histology Image Classification
• Deep Learning for Lung Cancer Detection: Tackling the Kaggle Data Science Bowl 2017 Challenge
• Cancer image classification based on DenseNet model
• Discriminative Pattern Mining for Breast Cancer Histopathology Image Classification via Fully Convolutional Autoencoder
• A Survey of Deep Learning for Scientific Discovery
• Leveraging Uncertainty Estimates for Predicting Segmentation Quality
• Deep Weakly-Supervised Learning Methods for Classification and Localization in Histology Images: A Survey
• Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learning
• Survey of XAI in digital pathology
• Interpreting a Recurrent Neural Network's Predictions of ICU Mortality Risk
• Terabyte-scale Deep Multiple Instance Learning for Classification and Localization in Pathology
• Semi-Supervised Histology Classification using Deep Multiple Instance Learning and Contrastive Predictive Coding
• Operational Calibration: Debugging Confidence Errors for DNNs in the Field
• A Dual-branch Self-supervised Representation Learning Framework for Tumour Segmentation in Whole Slide Images
• CELNet: Evidence Localization for Pathology Images using Weakly Supervised Learning
• Supervised Transfer Learning at Scale for Medical Imaging
• CAMEL: A Weakly Supervised Learning Framework for Histopathology Image Segmentation
• A Case for the Score: Identifying Image Anomalies using Variational Autoencoder Gradients
• Multi-Scale Hybrid Vision Transformer for Learning Gastric Histology: AI-Based Decision Support System for Gastric Cancer Treatment
• Yottixel -- An Image Search Engine for Large Archives of Histopathology Whole Slide Images
• Increasing a microscope's effective field of view via overlapped imaging and machine learning
• Representation Learning for Electronic Health Records
• Multimodal Multitask Representation Learning for Pathology Biobank Metadata Prediction
• Evaluating histopathology transfer learning with ChampKit
• Sample Efficient Semantic Segmentation using Rotation Equivariant Convolutional Networks
• Self-supervised learning-based general laboratory progress pretrained model for cardiovascular event detection
• Fine-Tuning and Training of DenseNet for Histopathology Image Representation Using TCGA Diagnostic Slides
• Evolved Explainable Classifications for Lymph Node Metastases
• ConvPath: A Software Tool for Lung Adenocarcinoma Digital Pathological Image Analysis Aided by Convolutional Neural Network
• Generalizing multistain immunohistochemistry tissue segmentation using one-shot color deconvolution deep neural networks
• Deep Learning Models for Digital Pathology
• Deep Learning Framework for Multi-class Breast Cancer Histology Image Classification
• Automatic breast cancer grading in lymph nodes using a deep neural network
• PseudoEdgeNet: Nuclei Segmentation only with Point Annotations
• A Robust and Effective Approach Towards Accurate Metastasis Detection and pN-stage Classification in Breast Cancer
• Multi-Stage Pathological Image Classification using Semantic Segmentation
• Using Apple Machine Learning Algorithms to Detect and Subclassify Non-Small Cell Lung Cancer
• It's easy to fool yourself: Case studies on identifying bias and confounding in bio-medical datasets
• Blind deblurring for microscopic pathology images using deep learning networks
• The 2ST-UNet for Pneumothorax Segmentation in Chest X-Rays using ResNet34 as a Backbone for U-Net
• Reversed Active Learning based Atrous DenseNet for Pathological Image Classification
• PFA-ScanNet: Pyramidal Feature Aggregation with Synergistic Learning for Breast Cancer Metastasis Analysis
• Dual-stream Maximum Self-attention Multi-instance Learning
• Natural vs Balanced Distribution in Deep Learning on Whole Slide Images for Cancer Detection
• A deep learning based multiscale approach to segment cancer area in liver whole slide image
• Transformation Invariant Cancerous Tissue Classification Using Spatially Transformed DenseNet
• An XAI Approach to Deep Learning Models in the Detection of DCIS
• Meta-SVDD: Probabilistic Meta-Learning for One-Class Classification in Cancer Histology Images
• A Computer-Aided Diagnosis System for Breast Pathology: A Deep Learning Approach with Model Interpretability from Pathological Perspective
• PointNu-Net: Keypoint-assisted Convolutional Neural Network for Simultaneous Multi-tissue Histology Nuclei Segmentation and Classification
• P3SGD: Patient Privacy Preserving SGD for Regularizing Deep CNNs in Pathological Image Classification
• DeepIso: A Deep Learning Model for Peptide Feature Detection
• Human-centric Metric for Accelerating Pathology Reports Annotation
• Intensity augmentation for domain transfer of whole breast segmentation in MRI
• Rotation Equivariant CNNs for Digital Pathology
• Detection and Classification of Breast Cancer Metastates Based on U-Net
• Chameleon: A Semi-AutoML framework targeting quick and scalable development and deployment of production-ready ML systems for SMEs
• Label Super Resolution with Inter-Instance Loss
• An Interpretable Algorithm for Uveal Melanoma Subtyping from Whole Slide Cytology Images
• Understanding the Mechanism of Deep Learning Framework for Lesion Detection in Pathological Images with Breast Cancer
• Analysis of Big Data Technology for Health Care Services
• A machine learning method correlating pulse pressure wave data with pregnancy
• Deep Learning for Estimating Synaptic Health of Primary Neuronal Cell Culture
• Large-scale Gastric Cancer Screening and Localization Using Multi-task Deep Neural Network
• Deep Generative Classifiers for Thoracic Disease Diagnosis with Chest X-ray Images
• Context-encoding Variational Autoencoder for Unsupervised Anomaly Detection -- Short Paper
• Effects of annotation granularity in deep learning models for histopathological images