TY - GEN
T1 - RWKVFUSE
T2 - 23rd IEEE International Symposium on Biomedical Imaging, ISBI 2026
AU - Al-Qurri, Ahmed
AU - Sabih, Mohammad
AU - Almekkawy, Mohamed
N1 - Publisher Copyright:
© 2026 IEEE.
PY - 2026
Y1 - 2026
N2 - Medical image segmentation is essential for accurate lesion identification and disease diagnosis. Conventional CNN-based methods are limited by their restricted receptive fields, whereas Transformer models effectively capture global dependencies but incur quadratic computational costs. Recently, sequence modeling approaches with linear attention mechanisms, such as the RWKV, have demonstrated the ability to model long-range dependencies with linear complexity. To address this, we propose a dual-encoder architecture, RWKVFuse, which integrates VRWKV and CNNs to extract fine-grained spatial features and long-range dependencies. Extensive experiments across four imaging modalities show that the proposed RWKVFuse model outperforms state-of-the-art methods in metrics such as DSC and HD95. Our code is available at: https://github.com/alqurri/rwkv.
AB - Medical image segmentation is essential for accurate lesion identification and disease diagnosis. Conventional CNN-based methods are limited by their restricted receptive fields, whereas Transformer models effectively capture global dependencies but incur quadratic computational costs. Recently, sequence modeling approaches with linear attention mechanisms, such as the RWKV, have demonstrated the ability to model long-range dependencies with linear complexity. To address this, we propose a dual-encoder architecture, RWKVFuse, which integrates VRWKV and CNNs to extract fine-grained spatial features and long-range dependencies. Extensive experiments across four imaging modalities show that the proposed RWKVFuse model outperforms state-of-the-art methods in metrics such as DSC and HD95. Our code is available at: https://github.com/alqurri/rwkv.
UR - https://www.scopus.com/pages/publications/105041663631
UR - https://www.scopus.com/pages/publications/105041663631#tab=citedBy
U2 - 10.1109/ISBI61048.2026.11515577
DO - 10.1109/ISBI61048.2026.11515577
M3 - Conference contribution
AN - SCOPUS:105041663631
T3 - Proceedings - International Symposium on Biomedical Imaging
BT - ISBI 2026 - 23rd IEEE International Symposium on Biomedical Imaging
PB - IEEE Computer Society
Y2 - 8 April 2026 through 11 April 2026
ER -