Multimodal attention-enhanced network of segmenting acute ischemic stroke from perfusion images

Acute ischemic stroke (AIS) is a major cause of long-term disability and mortality worldwide. Accurate segmentation of stroke lesions, particularly the infarct core and penumbra, is critical for effective treatment planning. In this study, we propose a deep learning-based multimodal segmentation network to improve the accuracy of AIS lesion delineation. The proposed framework consists of independent encoders, Multimodal Spatial and Channel Fusion (MSCF) modules, and Decoder Spatial and Channel Attention (DSCA) modules. Independent encoders are used to extract modality-specific features, while the MSCF module integrates complementary information across modalities. The DSCA module is introduced in the decoder to refine feature fusion between encoder representations and decoder features. The model was trained and evaluated on the ISLES SPES 2015, ISLES 2017, and ISLES 2018 datasets using Dice Similarity Coefficient (DSC), Hausdorff Distance (HD95), Recall, and Precision. On the ISLES SPES 2015 dataset, the proposed method achieved a DSC of 83.91%, HD95 of 4.16 mm, Recall of 78.23%, and Precision of 88.63%. On the ISLES 2017 and ISLES 2018 datasets, the model achieved DSC values of 41.32% and 54.90%, with HD95 values of 25.10 mm and 23.06 mm, respectively. These results demonstrate that the proposed multimodal fusion strategy effectively improves segmentation accuracy and robustness for AIS lesion segmentation.