Uncertainty Awareness Enables Efficient Labeling for Cancer Subtyping in Digital Pathology

TL;DR

This paper introduces an uncertainty-aware self-supervised learning approach for cancer subtyping in digital pathology, enabling efficient annotation and achieving state-of-the-art results with minimal labeled data.

Contribution

It integrates uncertainty estimation into contrastive learning to selectively annotate crucial images, reducing labeling effort and improving classification performance.

Findings

Achieves state-of-the-art cancer subtyping with only 1-10% annotations.

Uses uncertainty scores to guide selective labeling.

Enhances efficiency and precision in digital pathology classification.

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 improves 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.