The effect of variable labels on deep learning models trained to predict breast density

TL;DR

This study investigates how variability in expert-assessed breast density labels impacts the performance and internal representations of deep learning models used for density prediction, highlighting the importance of label consistency.

Contribution

It demonstrates that label variability significantly affects the mapping from model representations to density predictions, but has limited impact on the learned representations themselves.

Findings

Variability in density labels alters the mapping from representations to predictions.

Removing label distribution variation improves correlation coefficients from 0.751 to over 0.815.

Model representations remain stable despite label variability, with no significant difference in correlation.

Abstract

Purpose: High breast density is associated with reduced efficacy of mammographic screening and increased risk of developing breast cancer. Accurate and reliable automated density estimates can be used for direct risk prediction and passing density related information to further predictive models. Expert reader assessments of density show a strong relationship to cancer risk but also inter-reader variation. The effect of label variability on model performance is important when considering how to utilise automated methods for both research and clinical purposes. Methods: We utilise subsets of images with density labels to train a deep transfer learning model which is used to assess how label variability affects the mapping from representation to prediction. We then create two end-to-end deep learning models which allow us to investigate the effect of label variability on the model representation formed. Results: We show that the trained mappings from representations to labels are altered considerably by the variability of reader scores. Training on labels with distribution variation removed causes the Spearman rank correlation coefficients to rise from $0.751\pm0.002$ to either $0.815\pm0.006$ when averaging across readers or $0.844\pm0.002$ when averaging across images. However, when we train different models to investigate the representation effect we see little difference, with Spearman rank correlation coefficients of $0.846\pm0.006$ and $0.850\pm0.006$ showing no statistically significant difference in the quality of the model representation with regard to density prediction. Conclusions: We show that the mapping between representation and mammographic density prediction is significantly affected by label variability. However, the effect of the label variability on the model representation is limited.