analiza aiqbench
DualMiT-Net introduces a novel dual-branch architecture combining Mix Transformers for local mass details with EfficientNet encoders for global breast context, achieving superior segmentation metrics on the CBIS-DDSM dataset. This approach directly addresses challenges in mammography such as low contrast and irregular boundaries by fusing localized texture information with wider tissue surroundings via a spatial gating mechanism. The study demonstrates that integrating these distinct feature representations yields more accurate mass delineation than standard encoder-decoder baselines, advancing computer-aided diagnosis capabilities.
DualMiT-Net: Local-Global Transformer-Convolutional Fusion for Breast Mass Segmentation in Mammographic Regions of Interest
arXiv:2608.15019v1 Announce Type: cross Abstract: Breast mass segmentation is an important step in computer-aided mammography, but it remains difficult because masses can have low contrast, irregular shapes, and boundaries that blend with surrounding breast tissue. To address this problem, we present DualMiT-Net, a dual-branch network that uses both a focused view of the mass and a wider view of the surrounding tissue. The local branch uses a Mix Transformer (MiT-B5) encoder to learn mass shape, texture, and boundary information, while the global branch uses an EfficientNet-B5 encoder to learn surrounding breast context. Features from the two branches are shared at the deeper encoder levels and are then progressively fused in a single decoder. A spatial gate controls how much global information is added during decoding. We also evaluated four input representations and selected a percentile-windowed mammogram combined with a Gabor texture response. The model was trained and evaluated on the mass subset of the Curated Breast Imaging Subset of the Digital Database for Screening Mammography (CBIS-DDSM) using a patient-level split. Across three training runs, DualMiT-Net with exponential moving average weights achieved a mean Dice coefficient of 0.9375 and a mean Intersection over Union of 0.8834. It also achieved better Dice and IoU scores than six standard encoder-decoder baselines trained using the same data and training settings. These results show that combining local mass information with wider breast context can provide accurate and consistent breast mass segmentation.
Abstract:Breast mass segmentation is an important step in computer-aided mammography, but it remains difficult because masses can have low contrast, irregular shapes, and boundaries that blend with surrounding breast tissue. To address this problem, we present DualMiT-Net, a dual-branch network that uses both a focused view of the mass and a wider view of the surrounding tissue. The local branch uses a Mix Transformer (MiT-B5) encoder to learn mass shape, texture, and boundary information, while the global branch uses an EfficientNet-B5 encoder to learn surrounding breast context. Features from the two branches are shared at the deeper encoder levels and are then progressively fused in a single decoder. A spatial gate controls how much global information is added during decoding. We also evaluated four input representations and selected a percentile-windowed mammogram combined with a Gabor texture response. The model was trained and evaluated on the mass subset of the Curated Breast Imaging Subset of the Digital Database for Screening Mammography (CBIS-DDSM) using a patient-level split. Across three training runs, DualMiT-Net with exponential moving average weights achieved a mean Dice coefficient of 0.9375 and a mean Intersection over Union of 0.8834. It also achieved better Dice and IoU scores than six standard encoder-decoder baselines trained using the same data and training settings. These results show that combining local mass information with wider breast context can provide accurate and consistent breast mass segmentation.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2608.15019 [cs.CV]
(or arXiv:2608.15019v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2608.15019 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 334 cuvinte
Subiecte conexe
News
Quantum Computing Report · 25 iul. 2026
Multimodal Surface EMG Hand Gesture Recognition Using Query-Based Transformers for Prosthetic Control
arXiv · 28 iul. 2026
HyBDM: Multi-Scale Hybrid Experts for Time Series Forecasting with Bidirectional Dependency Modeling
arXiv · 21 iul. 2026
Problems of Proliferation
LessWrong · 29 sept. 2026