Mitigating annotation shift in cancer classification using single image generative models

TL;DR

This paper addresses annotation shift challenges in breast cancer classification by using single-image generative models for data augmentation, significantly improving model robustness with minimal additional annotations.

Contribution

It introduces a novel data augmentation method using single-image generative models to mitigate annotation shift in breast cancer classification tasks.

Findings

Annotation shift significantly impacts multiclass classification performance.

Using as few as four in-domain annotations, the proposed augmentation reduces annotation shift effects.

Ensemble models trained with different augmentation regimes further enhance classification accuracy.

Abstract

Artificial Intelligence (AI) has emerged as a valuable tool for assisting radiologists in breast cancer detection and diagnosis. However, the success of AI applications in this domain is restricted by the quantity and quality of available data, posing challenges due to limited and costly data annotation procedures that often lead to annotation shifts. This study simulates, analyses and mitigates annotation shifts in cancer classification in the breast mammography domain. First, a high-accuracy cancer risk prediction model is developed, which effectively distinguishes benign from malignant lesions. Next, model performance is used to quantify the impact of annotation shift. We uncover a substantial impact of annotation shift on multiclass classification performance particularly for malignant lesions. We thus propose a training data augmentation approach based on single-image generative models for the affected class, requiring as few as four in-domain annotations to considerably mitigate annotation shift, while also addressing dataset imbalance. Lastly, we further increase performance by proposing and validating an ensemble architecture based on multiple models trained under different data augmentation regimes. Our study offers key insights into annotation shift in deep learning breast cancer classification and explores the potential of single-image generative models to overcome domain shift challenges.