Topological Conditioning for Mammography Models via a Stable Wavelet-Persistence Vectorization

2025-12-10

AI Summary

In breast cancer screening, mammography models suffer from high false-negative and false-positive rates and poor generalizability across imaging devices and diverse populations. To address this, we propose a wavelet-enhanced persistent homology vectorization method that integrates stable, robust topological feature maps—insensitive to intensity perturbations—into a two-stage detection framework via channel-wise concatenation. This work is the first to introduce wavelet-persistent homology vectorization as an interpretable, unsupervised, structure-aware prior in medical image analysis, requiring no additional annotations. Our approach synergistically combines multi-scale wavelet decomposition with topological data analysis (TDA) and is implemented end-to-end within a ConvNeXt architecture. On the INbreast independent test set, patient-level AUC improves significantly from 0.55 to 0.75. Moreover, the method demonstrates superior cross-domain generalizability across three geographically distinct cohorts (U.S., Portugal, China) and dual imaging modalities (film and digital mammography).

Technology Category

• Computer Vision: Medical and Biological Imaging
• Machine Learning: Multi-instance/Multi-view Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Breast cancer is the most commonly diagnosed cancer in women and a leading cause of cancer death worldwide. Screening mammography reduces mortality, yet interpretation still suffers from substantial false negatives and false positives, and model accuracy often degrades when deployed across scanners, modalities, and patient populations. We propose a simple conditioning signal aimed at improving external performance based on a wavelet based vectorization of persistent homology. Using topological data analysis, we summarize image structure that persists across intensity thresholds and convert this information into spatial, multi scale maps that are provably stable to small intensity perturbations. These maps are integrated into a two stage detection pipeline through input level channel concatenation. The model is trained and validated on the CBIS DDSM digitized film mammography cohort from the United States and evaluated on two independent full field digital mammography cohorts from Portugal (INbreast) and China (CMMD), with performance reported at the patient level. On INbreast, augmenting ConvNeXt Tiny with wavelet persistence channels increases patient level AUC from 0.55 to 0.75 under a limited training budget.

Problem

• Improves mammography model accuracy across diverse scanners and populations
• Reduces false positives and negatives in breast cancer screening
• Enhances external performance using topological data analysis conditioning

Innovation

• Wavelet-based vectorization of persistent homology
• Stable spatial multi-scale maps from topology
• Two-stage detection with channel concatenation