mlsanityData to 7 Oct 2026

Whole-Slide Mitosis Detection in H&E Breast Histology Using PHH3 as a Reference to Train Distilled Stain-Invariant Convolutional Networks

arXiv 1808.05896 ↗First posted 17 Aug 2018cs.CVcited by 57

Not on the map. Why?

Abstract (arXiv)

Manual counting of mitotic tumor cells in tissue sections constitutes one of the strongest prognostic markers for breast cancer. This procedure, however, is time-consuming and error-prone. We developed a method to automatically detect mitotic figures in breast cancer tissue sections based on convolutional neural networks (CNNs). Application of CNNs to hematoxylin and eosin (H&E) stained histological tissue sections is hampered by: (1) noisy and expensive reference standards established by pathologists, (2) lack of generalization due to staining variation across laboratories, and (3) high computational requirements needed to process gigapixel whole-slide images (WSIs). In this paper, we present a method to train and evaluate CNNs to specifically solve these issues in the context of mitosis detection in breast cancer WSIs. First, by combining image analysis of mitotic activity in phosphohistone-H3 (PHH3) restained slides and registration, we built a reference standard for mitosis detection in entire H&E WSIs requiring minimal manual annotation effort. Second, we designed a data augmentation strategy that creates diverse and realistic H&E stain variations by modifying the hematoxylin and eosin color channels directly. Using it during training combined with network ensembling resulted in a stain invariant mitosis detector. Third, we applied knowledge distillation to reduce the computational requirements of the mitosis detection ensemble with a negligible loss of performance. The system was trained in a single-center cohort and evaluated in an independent multicenter cohort from The Cancer Genome Atlas on the three tasks of the Tumor Proliferation Assessment Challenge (TUPAC). We obtained a performance within the top-3 best methods for most of the tasks of the challenge.

Key facts. Cited by 57 papers in the map (data through 7 Oct 2026). First posted on arXiv 17 Aug 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 8; 2022: 13; 2023: 14; 2024: 11; 2025: 8; 2026: 3.

57 papers.

Sort:

By area:

  1. 2021Accuracy on the Line: On the Strong Correlation Between Out-of-Distribution and In-Distribution Generalizationcited by 248
  2. 2023Artificial Intelligence for Digital and Computational Pathologycited by 106
  3. 2021A Comprehensive Review of Computer-aided Whole-slide Image Analysis: from Datasets to Feature Extraction, Segmentation, Classification, and Detection Approachescited by 44
  4. 2021NuCLS: A scalable crowdsourcing, deep learning approach and dataset for nucleus classification, localization and segmentationcited by 33
  5. 2022RandStainNA: Learning Stain-Agnostic Features from Histology Slides by Bridging Stain Augmentation and Normalizationcited by 32
  6. 2023Domain Generalization in Computational Pathology: Survey and Guidelinescited by 29
  7. 2023Domain generalization across tumor types, laboratories, and species -- insights from the 2022 edition of the Mitosis Domain Generalization Challengecited by 25
  8. 2022AdvMIL: Adversarial Multiple Instance Learning for the Survival Analysis on Whole-Slide Imagescited by 24
  9. 2024Towards Large-Scale Training of Pathology Foundation Modelscited by 19
  10. 2022Test-time image-to-image translation ensembling improves out-of-distribution generalization in histopathologycited by 13
  11. 2021Comparison of Consecutive and Re-stained Sections for Image Registration in Histopathologycited by 12
  12. 2023Deep Learning in Breast Cancer Imaging: A Decade of Progress and Future Directionscited by 10
38 more
  1. 2021A State-of-the-art Survey of Artificial Neural Networks for Whole-slide Image Analysis:from Popular Convolutional Neural Networks to Potential Visual Transformerscited by 7
  2. 2022Deep Feature based Cross-slide Registrationcited by 7
  3. 2023Slideflow: Deep Learning for Digital Histopathology with Real-Time Whole-Slide Visualizationcited by 7
  4. 2024Connect Later: Improving Fine-tuning for Robustness with Targeted Augmentationscited by 7
  5. 2022Mitosis Detection, Fast and Slow: Robust and Efficient Detection of Mitotic Figurescited by 6
  6. 2023Assessing and Enhancing Robustness of Deep Learning Models with Corruption Emulation in Digital Pathologycited by 6
  7. 2022Breast cancer detection using artificial intelligence techniques: A systematic literature reviewcited by 5
  8. 2022Augment like there's no tomorrow: Consistently performing neural networks for medical imagingcited by 5
  9. 2022Detecting Mitoses with a Convolutional Neural Network for MIDOG 2022 Challengecited by 5
  10. 2024Self-supervised Vision Transformer are Scalable Generative Models for Domain Generalizationcited by 4
  11. 2024OMG-Net: A Deep Learning Framework Deploying Segment Anything to Detect Pan-Cancer Mitotic Figures from Haematoxylin and Eosin-Stained Slidescited by 4
  12. 2026Histopath-C: Towards Realistic Domain Shifts for Histopathology Vision-Language Adaptationcited by 4
  13. 2021MitoDet: Simple and robust mitosis detectioncited by 3
  14. 2022A review of machine learning approaches, challenges and prospects for computational tumor pathologycited by 3
  15. 2023Ranking Loss and Sequestering Learning for Reducing Image Search Bias in Histopathologycited by 3
  16. 2023Consistency Regularization for Domain Generalization with Logit Attribution Matchingcited by 3
  17. 2023Group Distributionally Robust Knowledge Distillationcited by 3
  18. 2022Visual attention analysis of pathologists examining whole slide images of Prostate cancercited by 2
  19. 2023Experts' cognition-driven ensemble deep learning for external validation of predicting pathological complete response to neoadjuvant chemotherapy from histological images in breast cancercited by 2
  20. 2024Scalable, Trustworthy Generative Model for Virtual Multi-Staining from H&E Whole Slide Imagescited by 2
  21. 2024SRA: A Novel Method to Improve Feature Embedding in Self-supervised Learning for Histopathological Imagescited by 2
  22. 2024Resource Efficient Multi-stain Kidney Glomeruli Segmentation via Self-supervisioncited by 2
  23. 2025From Traditional to Deep Learning Approaches in Whole Slide Image Registration: A Methodological Reviewcited by 2
  24. 2025Test-Time Adaptation by Causal Trimmingcited by 2
  25. 2021Fast whole-slide cartography in colon cancer histology using superpixels and CNN classificationcited by 1
  26. 2022Contrastive learning-based computational histopathology predict differential expression of cancer driver genescited by 1
  27. 2022Radial Prediction Domain Adaption Classifier for the MIDOG 2022 Challengecited by 1
  28. 2022Learning Melanocytic Cell Masks from Adjacent Stained Tissuecited by 1
  29. 2024On the Value of PHH3 for Mitotic Figure Detection on H&E-stained Imagescited by 1
  30. 2024Advancing Histopathology with Deep Learning Under Data Scarcity: A Decade in Reviewcited by 1
  31. 2025Improving Out-of-Domain Robustness with Targeted Augmentation in Frequency and Pixel Spacescited by 1
  32. 2026A Comprehensive Benchmark of Histopathology Foundation Models for Kidney Digital Pathology Imagescited by 1
  33. 2021Lung Cancer Lesion Detection in Histopathology Images Using Graph-Based Sparse PCA Networkcited by 0
  34. 2023Adaptive Domain Generalization for Digital Pathology Imagescited by 0
  35. 2023Rethinking Mitosis Detection: Towards Diverse Data and Feature Representationcited by 0
  36. 2023ALFA -- Leveraging All Levels of Feature Abstraction for Enhancing the Generalization of Histopathology Image Classification Across Unseen Hospitalscited by 0
  37. 2023Stain Consistency Learning: Handling Stain Variation for Automatic Digital Pathology Segmentationcited by 0
  38. 2024Immunocto: a massive immune cell database auto-generated for histopathologycited by 0

50 most cited of 57 All 57 (JSON)