Abstract #1715

Spatial-Temporal Mamba Network for Accurate Breast Tumor Segmentation in DCE-MRI

Heng Zhang¹, Meng Wang², Ya Ren², Jie Wen², Wei Cui², Binze Han³, Dehong Luo², Zhou Liu⁴, and Na Zhang^5,6

¹Faculty of Robot Science and Engineering, Northeastern University, Shenyang, China, ²Department of Radiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital & Shenzhen Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Shenzhen, China, ³Southern University of Science and Technology (SUSTech), Shenzhen, China, ⁴National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital & Shenzhen Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Shenzhen, China, ⁵Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China, ⁶Key Laboratory of Biomedical Imaging Science and System, Chinese Academy of Sciences, Shenzhen, China

Synopsis

Keywords: Analysis/Processing, Machine Learning/Artificial Intelligence

Motivation: Breast tumor segmentation with DCE-MRI facilitates downstream characterization, diagnosis, and prognostication. However, current methods often overlook the temporal hemodynamic information inherent in DCE-MRI, limiting segmentation accuracy. This study addresses this gap by integrating spatial and temporal data to enhance tumor segmentation.

Goal(s): To develop a model that captures both spatial and temporal features in DCE-MRI to enhance accuracy.

Approach: We propose a Spatial-Temporal Mamba Network, integrating 3D spatial structures and multi-phase hemodynamic features using a 4D encoder and specialized modules for both spatial and temporal feature extraction.

Results: Our model achieved superior performance in DSC and HD metrics compared to state-of-the-art methods.

Impact: Our results show that the proposed model can significantly improve tumor segmentation accuracy in DCE-MRI by utilizing both spatial and temporal features. This advancement holds promise for more accurate breast cancer diagnosis and better-informed treatment planning.

How to access this content:

For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.

After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.

After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.

Click here for more information on becoming a member.