# Hybrid Supervision Learning for Pathology Whole Slide Image Classification

arXiv: https://arxiv.org/abs/2107.00934 · HTML: https://mlsanity.com/paper/2107.00934/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 27 references that are in the map.
- First posted on arXiv 2 Jul 2021.

## Builds on (most cited first, 15 of 15)

- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [Deep Layer Aggregation](https://mlsanity.com/paper/1707.06484/) (2017)
- [Patch-based Convolutional Neural Network for Whole Slide Tissue Image Classification](https://mlsanity.com/paper/1504.07947/) (2015)
- [Learning to Segment Every Thing](https://mlsanity.com/paper/1711.10370/) (2017)
- [CGC-Net: Cell Graph Convolutional Network for Grading of Colorectal Cancer Histology Images](https://mlsanity.com/paper/1909.01068/) (2019)
- [Thoracic Disease Identification and Localization with Limited Supervision](https://mlsanity.com/paper/1711.06373/) (2017)
- [A Generalized Deep Learning Framework for Whole-Slide Image Segmentation and Analysis](https://mlsanity.com/paper/2001.00258/) (2020)
- [Context-Aware Convolutional Neural Network for Grading of Colorectal Cancer Histology Images](https://mlsanity.com/paper/1907.09478/) (2019)
- [Development and Validation of a Deep Learning Algorithm for Improving Gleason Scoring of Prostate Cancer](https://mlsanity.com/paper/1811.06497/) (2018)
- [Signet Ring Cell Detection With a Semi-supervised Learning Framework](https://mlsanity.com/paper/1907.03954/) (2019)
- [Deep Learning with Mixed Supervision for Brain Tumor Segmentation](https://mlsanity.com/paper/1812.04571/) (2018)
- [Multi-Stage Pathological Image Classification using Semantic Segmentation](https://mlsanity.com/paper/1910.04473/) (2019)
- [HybridNet: Classification and Reconstruction Cooperation for Semi-Supervised Learning](https://mlsanity.com/paper/1807.11407/) (2018)
- [Pseudo Mask Augmented Object Detection](https://mlsanity.com/paper/1803.05858/) (2018)
- [Pathologist-Level Grading of Prostate Biopsies with Artificial Intelligence](https://mlsanity.com/paper/1907.01368/) (2019)

## Built on by (most cited first, 5 of 5)

- [Deep Learning for Computational Cytology: A Survey](https://mlsanity.com/paper/2202.05126/) (2022)
- [AEM: Attention Entropy Maximization for Multiple Instance Learning based Whole Slide Image Classification](https://mlsanity.com/paper/2406.15303/) (2024)
- [Learning with Noisy Labels by Adaptive Gradient-Based Outlier Removal](https://mlsanity.com/paper/2306.04502/) (2023)
- [ULF: Unsupervised Labeling Function Correction using Cross-Validation for Weak Supervision](https://mlsanity.com/paper/2204.06863/) (2022)
- [LESS: Label-efficient Multi-scale Learning for Cytological Whole Slide Image Screening](https://mlsanity.com/paper/2306.03407/) (2023)

## Abstract (arXiv)

Weak supervision learning on classification labels has demonstrated high performance in various tasks, while a few pixel-level fine annotations are also affordable. Naturally a question comes to us that whether the combination of pixel-level (e.g., segmentation) and image level (e.g., classification) annotation can introduce further improvement. However in computational pathology this is a difficult task for this reason: High resolution of whole slide images makes it difficult to do end-to-end classification model training, which is challenging to research of weak or hybrid supervision learning in the past. To handle this problem, we propose a hybrid supervision learning framework for this kind of high resolution images with sufficient image-level coarse annotations and a few pixel-level fine labels. This framework, when applied in training patch model, can carefully make use of coarse image-level labels to refine generated pixel-level pseudo labels. Complete strategy is proposed to suppress pixel-level false positives and false negatives. A large hybrid annotated dataset is used to evaluate the effectiveness of hybrid supervision learning. By extracting pixel-level pseudo labels in initially image-level labeled samples, we achieve 5.2% higher specificity than purely training on existing labels while retaining 100% sensitivity, in the task of image-level classification to be positive or negative.