ReaderAdaptNet: modeling reader variability in breast imaging with reader-specific embeddings

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: Breast density: 8.0 percentage point improvement (76.4% to 84.4%); BPE: 7.0 percentage point improvement (65.1% to 72.1%)

• Primary Outcome: Classification accuracy improvement for breast imaging interpretation tasks

• Nnt Or Sensitivity: Relative improvement: 10.5% for breast density, 10.8% for BPE classification

• Confidence Interval: Not reported

Clinical Application

High feasibility due to parameter-efficient design and ability to adapt to new readers with minimal additional training data Highly relevant to Australian breast screening programs (BreastScreen Australia) where standardizing reader interpretation could improve screening accuracy and reduce false positives/negatives Radiologists and healthcare institutions performing breast imaging interpretation, particularly mammography and contrast-enhanced mammography

Abstract

Objective.Inter-reader variability remains a major challenge in breast imaging interpretation, particularly for ordinal classification tasks such as breast density and background parenchymal enhancement (BPE). These visual assessments are prone to inconsistencies due to subjective interpretation, limiting the reliability of AI-based models trained on aggregated or noisy labels. This work aims to account for this variability rather than collapse it.Approach.We propose ReaderAdaptNet, a reader-adaptive network that explicitly models inter-reader variability through reader-specific embeddings. A novel two-stage deep learning framework underpins the training process of the model. The first stage jointly learns discriminative image features and low-dimensional embeddings representing individual annotation styles, enabling personalized classification. The second stage allows for embedding calibration using a small set of labeled examples, enabling rapid adaptation to new readers or consensus definitions without retraining the full model. The method is designed for multi-reader datasets and is evaluated on two breast classification tasks: breast density classification from full-field digital mammography and BPE classification from contrast-enhanced mammography.Main results.Results demonstrate that reader embeddings improve both individual and consensus-level performance, and calibrated embeddings enable flexible, low-cost personalization. Reader-specific embeddings of dimension 32 improved the mean classification accuracy across readers, increasing from 76.4% to 84.4% for breast density, and from 65.1% to 72.1% for BPE, compared to the proposed baseline method.Significance.By disentangling stable image features from reader-specific decision tendencies, ReaderAdaptNet provides a parameter-efficient, interpretable route topersonalization(alignment to a given reader) orunification(alignment to an institutional standard) under real-world variability.