Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
The paper introduces MC-GenRef, a framework for microcalcification segmentation in mammograms that avoids the need for real dense annotations by using synthetic data generation and test-time generative posterior refinement (TT-GPR). They generate realistic synthetic MC data using an image formation model and train a base segmentor and a rectified-flow generator. TT-GPR then refines segmentation at inference by treating it as approximate posterior inference, using seed-consistent rectified-flow projections to create surrogate targets and iteratively refine logits with regularization, improving performance and robustness, especially in miss-sensitive metrics.
Forget painstakingly labeling mammograms: this method uses synthetic data and generative refinement to segment microcalcifications with surprising accuracy, even across different datasets.
Microcalcification (MC) analysis is clinically important in screening mammography because clustered puncta can be an early sign of malignancy, yet dense MC segmentation remains challenging: targets are extremely small and sparse, dense pixel-level labels are expensive and ambiguous, and cross-site shift often induces texture-driven false positives and missed puncta in dense tissue. We propose MC-GenRef, a real dense-label-free framework that combines high-fidelity synthetic supervision with test-time generative posterior refinement (TT-GPR). During training, real negative mammogram patches are used as backgrounds, and physically plausible MC patterns are injected through a lightweight image formation model with local contrast modulation and blur, yielding exact image-mask pairs without real dense annotation. Using only these synthetic labeled pairs, MC-GenRef trains a base segmentor and a seed-conditioned rectified-flow (RF) generator that serves as a controllable generative prior. During inference, TT-GPR treats segmentation as approximate posterior inference: it derives a sparse seed from the current prediction, forms seed-consistent RF projections, converts them into case-specific surrogate targets through the frozen segmentor, and iteratively refines the logits with overlap-consistent and edge-aware regularization. On INbreast, the synthetic-only initializer achieved the best Dice without real dense annotations, while TT-GPR improved miss-sensitive performance to Recall and FNR, with strong class-balanced behavior (Bal.Acc., G-Mean). On an external private Yonsei cohort ( n=50 ), TT-GPR consistently improved the synthetic-only initializer under cross-site shift, increasing Dice and Recall while reducing FNR. These results suggest that test-time generative posterior refinement is a practical route to reduce MC misses and improve robustness without additional real dense labeling.