CVFSNet: A Cross View Fusion Scoring Network for end-to-end mTICI scoring

Abstract

The modified Thrombolysis In Cerebral Infarction (mTICI) score serves as one of the key clinical indicators to assess the success of the Mechanical Thrombectomy (MT), requiring physicians to inspect Digital Subtraction Angiography (DSA) images in both the coronal and sagittal views. However, assessing mTICI scores manually is time-consuming and has considerable observer variability. An automatic, objective, and end-to-end method for assigning mTICI scores may effectively avoid observer errors. Therefore, in this paper, we propose a novel Cross View Fusion Scoring Network (CVFSNet) for automatic, objective, and end-to-end mTICI scoring, which employs dual branches to simultaneously extract spatial–temporal features from coronal and sagittal views. Then, a novel Cross View Fusion Module (CVFM) is introduced to fuse the features from two views, which explores the positional characteristics of coronal and sagittal views to generate a pseudo-oblique sagittal feature and ultimately constructs more representative features to enhance the scoring performance. In addition, we provide AmTICIS, a newly collected and the first publicly available DSA image dataset with expert annotations for automatic mTICI scoring, which can effectively promote researchers to conduct studies of ischemic stroke based on DSA images and finally help patients get better medical treatment. Extensive experimentation results demonstrate the promising performance of our methods and the validity of the cross-view fusion module. Code and data will be available at https://github.com/xwjBupt/CVFSNet.