Multi-view DBT grid-attention detection framework

Xuan Liu; Yinhao Ren; Zisheng Liang; Lars J. Grimm; Jun Ge; Joseph Y. Lo

doi:10.1117/12.2653708

7 April 2023 Multi-view DBT grid-attention detection framework

Xuan Liu, Yinhao Ren, Zisheng Liang, Lars J. Grimm, Jun Ge, Joseph Y. Lo

Proceedings Volume 12465, Medical Imaging 2023: Computer-Aided Diagnosis; 124650I (2023) https://doi.org/10.1117/12.2653708
Event: SPIE Medical Imaging, 2023, San Diego, California, United States

Abstract

Most of the existing CAD frameworks for digital breast tomosynthesis (DBT) are single-view only, while radiologists typically utilize information from multiple screening views to better detect breast cancer lesions. Previously, we developed the Retina-Match framework for lesion detection that performed ipsilateral matching between CC and MLO views of the same breast. In this work, we improve that framework in both sampling strategy and feature extraction processes. We proposed a “hard negative” sampling strategy to train on more difficult ipsilateral lesion pairs to increase the robustness of the matching model. We introduced a grid-attention (GA) module to apply spatial attention mechanism for ipsilateral patch similarity learning. A screening DBT dataset with 4182 cases including 1498 (36%) cancers were used for training and testing. Case specificity improved by 15% at 96% sensitivity and average False Positive numbers per detection image (FPPI) decreased from 0.6 to 0.5 at 95% case sensitivity. These experiments indicated that proper ipsilateral matching result is the key to improve performance of multi-view lesion detection framework.

Conference Presentation

Citation Download Citation

Xuan Liu, Yinhao Ren, Zisheng Liang, Lars J. Grimm, Jun Ge, and Joseph Y. Lo "Multi-view DBT grid-attention detection framework", Proc. SPIE 12465, Medical Imaging 2023: Computer-Aided Diagnosis, 124650I (7 April 2023); https://doi.org/10.1117/12.2653708

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available