Learn like a Pathologist: Curriculum Learning by Annotator Agreement for Histopathology Image Classification

@article{Wei2021LearnLA,
  title={Learn like a Pathologist: Curriculum Learning by Annotator Agreement for Histopathology Image Classification},
  author={Jerry W. Wei and Arief A. Suriawinata and Bing Ren and Xiaoying Liu and Mikhail Lisovsky and Louis J. Vaickus and Charles Brown and Michael Baker and Mustafa Nasir-Moin and Naofumi Tomita and Lorenzo Torresani and Jason Wei and Saeed Hassanpour},
  journal={2021 IEEE Winter Conference on Applications of Computer Vision (WACV)},
  year={2021},
  pages={2472-2482}
}
Applying curriculum learning requires both a range of difficulty in data and a method for determining the difficulty of examples. In many tasks, however, satisfying these requirements can be a formidable challenge.In this paper, we contend that histopathology image classification is a compelling use case for curriculum learning. Based on the nature of histopathology images, a range of difficulty inherently exists among examples, and, since medical datasets are often labeled by multiple… 

Figures and Tables from this paper

Calibrating Histopathology Image Classifiers using Label Smoothing
TLDR
This paper proposes several label smoothing methods that utilize per-image annotator agreement and finds that they substantially improve model calibration, while maintaining (or even improving) accuracy.
Improving Self-supervised Learning with Hardness-aware Dynamic Curriculum Learning: An Application to Digital Pathology
TLDR
This paper dynamically leverage progressive harder examples via easy- to-hard and hard-to-very-hard samples during mini-batch downstream fine-tuning to improve the robustness and generalizability of SSL, and empirically shows that the approach is more generic and adaptable to any SSL methods and does not impose any additional overhead complexity.
A Survey on Domain Knowledge Powered Deep Learning for Medical Image Analysis
TLDR
This survey summarizes the current progress on introducing medical domain knowledge in deep learning models for various tasks like disease diagnosis, lesion, organ and abnormality detection, lesions and organ segmentation, and systematically categorizes different kinds of medicaldomain knowledge that have been utilized and the corresponding integrating methods.
Medical knowledge-guided deep curriculum learning for elbow fracture diagnosis from x-ray images
TLDR
This work proposes a novel deep learning method to diagnose elbow fracture from elbow X-ray images by integrating domain-specific medical knowledge into a curriculum learning framework and proposes an algorithm that updates the sampling probabilities at each epoch, applicable to other sampling-based curriculum learning frameworks.
Deep Learning Application for Analyzing of Constituents and Their Correlations in the Interpretations of Medical Images
TLDR
The novelty in this paper consists primarily in the unitary approach, of the constituent elements of DL models, namely, data, tools used by DL architectures or specifically constructed DL architecture combinations and highlighting their “key” features, for completion of tasks in current applications in the interpretation of medical images.
Predicting Oncogene Mutations of Lung Cancer Using Deep Learning and Histopathologic Features on Whole-Slide Images
TLDR
It is demonstrated that H&E stained FFPE LUAD whole slides could be utilized to predict oncogene mutations, such as EGFR, indicating that somatic mutations could present subtle morphological characteristics in histology slides, where deep learning-based feature extractors can learn such latent information.
Curriculum Learning: A Survey
TLDR
This survey shows how limits have been tackled in the literature, and presents curriculum learning instantiations for various tasks in machine learning, and constructs a multi-perspective clustering algorithm, linking the discovered clusters with the taxonomy.
On the Efficiency of Subclass Knowledge Distillation in Classification Tasks
TLDR
This work introduces a novel knowledge distillation framework for classification tasks where information on existing subclasses is available and taken into consideration, and measures how many extra information bits the teacher can provide for the student via the proposed Subclass Knowledge Distillation (SKD) framework.
LeRaC: Learning Rate Curriculum
Most curriculum learning methods require an approach to sort the data samples by difficulty, which is often cumbersome to perform. In this work, we propose a novel curriculum learning approach termed
...
1
2
...

References

SHOWING 1-10 OF 51 REFERENCES
Training Medical Image Analysis Systems like Radiologists
TLDR
The use of meta-training that models a classifier based on a series of tasks is explored, which is shown to be the best in the field for that application, compared to state of art baseline approaches: DenseNet, multiple instance learning and multi-task learning.
Deep Learning for Classification of Colorectal Polyps on Whole-slide Images
TLDR
An automatic image analysis method that can accurately classify different types of colorectal polyps on whole-slide images to help pathologists with this characterization and diagnosis and can reduce the cognitive burden on pathologists and improve their efficacy in histopathological characterization and follow-up recommendations.
Pathologist-level interpretable whole-slide cancer diagnosis with deep learning
TLDR
A novel pathology whole-slide diagnosis method, powered by artificial intelligence, to address the lack of interpretable diagnosis, which provides an innovative and reliable means for making diagnostic suggestions and can be deployed at low cost as next-generation, artificial intelligence-enhanced CAD technology for use in diagnostic pathology.
Pathologist-level classification of histologic patterns on resected lung adenocarcinoma slides with deep neural networks
TLDR
A deep learning model is proposed that automatically classifies the histologic patterns of lung adenocarcinoma on surgical resection slides by automatically pre-screening and highlighting cancerous regions prior to review.
Automated Gleason grading of prostate cancer tissue microarrays via deep learning
TLDR
A deep learning approach for automated Gleason grading of prostate cancer tissue microarrays with Hematoxylin and Eosin (H&E) staining achieves pathology expert-level stratification of patients into prognostically distinct groups, on the basis of disease-specific survival data available for the test cohort.
CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison
TLDR
A labeler is designed to automatically detect the presence of 14 observations in radiology reports, capturing uncertainties inherent in radiograph interpretation, in CheXpert, a large dataset that contains 224,316 chest radiographs of 65,240 patients.
Classification and mutation prediction from non–small cell lung cancer histopathology images using deep learning
TLDR
A deep convolutional neural network model is trained on whole-slide images obtained from The Cancer Genome Atlas to accurately and automatically classify them into LUAD, LUSC or normal lung tissue and predicts the ten most commonly mutated genes in LUAD.
CASED: Curriculum Adaptive Sampling for Extreme Data Imbalance
TLDR
The CASED learning framework makes no assumptions with regard to imaging modality or segmentation target and should generalize to other medical imaging problems where class imbalance is a persistent problem.
Weakly-supervised learning for lung carcinoma classification using deep learning
TLDR
A Convolution Neural Network based on the EfficientNet-B3 architecture is trained, using transfer learning and weakly-supervised learning, to predict carcinoma in Whole Slide Images using a training dataset of 3,554 WSIs.
...
1
2
3
4
5
...