Photometric Transformer Networks and Label Adjustment for Breast Density Prediction

TL;DR

This paper introduces novel deep learning techniques, including photometric transformer networks and label adjustment, to improve breast density prediction from mammograms by addressing normalization and grading variability issues.

Contribution

The paper proposes two innovative methods—adaptive photometric normalization and label distillation—that enhance prediction accuracy across multi-site mammogram datasets.

Findings

Significant performance improvements over previous methods.

Effective normalization of mammogram intensities.

Reduction in grading variability effects.

Abstract

Grading breast density is highly sensitive to normalization settings of digital mammogram as the density is tightly correlated with the distribution of pixel intensity. Also, the grade varies with readers due to uncertain grading criteria. These issues are inherent in the density assessment of digital mammography. They are problematic when designing a computer-aided prediction model for breast density and become worse if the data comes from multiple sites. In this paper, we proposed two novel deep learning techniques for breast density prediction: 1) photometric transformation which adaptively normalizes the input mammograms, and 2) label distillation which adjusts the label by using its output prediction. The photometric transformer network predicts optimal parameters for photometric transformation on the fly, learned jointly with the main prediction network. The label distillation, a type of pseudo-label techniques, is intended to mitigate the grading variation. We experimentally showed that the proposed methods are beneficial in terms of breast density prediction, resulting in significant performance improvement compared to various previous approaches.