MammoTracker: Mask-Guided Lesion Tracking in Temporal Mammograms

TL;DR

MammoTracker is a novel mask-guided framework for automated lesion tracking in temporal mammograms, improving accuracy and providing a large annotated dataset for breast cancer monitoring.

Contribution

The paper introduces MammoTracker, a new lesion tracking method with a coarse-to-fine approach and a large annotated dataset for training and evaluation.

Findings

Achieves 0.455 average overlap and 0.509 accuracy in lesion tracking.

Surpasses baseline models by 8% in performance.

Provides the largest dataset with over 20,000 lesion pairs for mammogram analysis.

Abstract

Accurate lesion tracking in temporal mammograms is essential for monitoring breast cancer progression and facilitating early diagnosis. However, automated lesion correspondence across exams remains a challenges in computer-aided diagnosis (CAD) systems, limiting their effectiveness. We propose MammoTracker, a mask-guided lesion tracking framework that automates lesion localization across consecutively exams. Our approach follows a coarse-to-fine strategy incorporating three key modules: global search, local search, and score refinement. To support large-scale training and evaluation, we introduce a new dataset with curated prior-exam annotations for 730 mass and calcification cases from the public EMBED mammogram dataset, yielding over 20000 lesion pairs, making it the largest known resource for temporal lesion tracking in mammograms. Experimental results demonstrate that MammoTracker achieves 0.455 average overlap and 0.509 accuracy, surpassing baseline models by 8%, highlighting its potential to enhance CAD-based lesion progression analysis. Our dataset will be available at https://gitlab.oit.duke.edu/railabs/LoGroup/mammotracker.