Uncertainty Awareness Enables Efficient Labeling for Cancer Subtyping in Digital Pathology

2025-02-26

IEEE Workshop/Winter Conference on Applications of Computer Vision

AI Summary

Cancer subtype classification in digital pathology faces challenges of high annotation costs and limited labeled data. To address this, we propose an uncertainty-aware self-supervised contrastive learning framework that, for the first time, integrates evidential deep learning into the contrastive learning pipeline. Our method dynamically generates evidence vectors to quantify predictive uncertainty, which guides an active learning strategy for selecting the most informative samples for iterative labeling. Evaluated on multiple benchmark datasets, our approach achieves state-of-the-art performance using only 1–10% of fully annotated data—outperforming fully supervised baselines while significantly improving both classification accuracy and annotation efficiency. The implementation is publicly available.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Learning & Optimization for CV
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Machine-learning-assisted cancer subtyping is a promising avenue in digital pathology. Cancer subtyping models however require careful training using expert annotations, so that they can be inferred with a degree of known certainty (or uncertainty). To this end, we introduce the concept of uncertainty awareness into a self-supervised contrastive learning model. This is achieved by computing an evidence vector at every epoch, which assesses the model's confidence in its predictions. The derived uncertainty score is then utilized as a metric to selectively label the most crucial images that require further annotation, thus iteratively refining the training process. With just 1-10% of strategically selected annotations, we attain state-of-the-art performance in cancer subtyping on benchmark datasets. Our method not only strategically guides the annotation process to minimize the need for extensive labeled datasets, but also improve the precision and efficiency of classifications. This development is particularly beneficial in settings where the availability of labeled data is limited, offering a promising direction for future research and application in digital pathology. Our code is available at https://github.com/Nirhoshan/AI-for-histopathology

Problem

• Improving cancer subtyping accuracy with limited expert annotations
• Reducing labeled data needs via strategic uncertainty-aware labeling
• Enhancing classification efficiency in digital pathology using self-supervised learning

Innovation

• Self-supervised contrastive learning with uncertainty awareness
• Evidence vectors compute model confidence per epoch
• Strategic annotation selection using uncertainty scores