MFSL-Net: A Modality Fusion and Shape Learning based Cascaded Network for Prostate Tumor Segmentation

Zhang, Fan; Zhang, Bo; Zhang, Zheng; Mi, Yue; Wu, Jingyun; Huang, Haiwen; Que, Xirong; Wang, Wendong

doi:10.1109/bigdata52589.2021.9671960

Cited by 2 publications

(1 citation statement)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4, respectively. These methods can be roughly divided into three categories: 1) universal segmentation methods, including 2D nnUnet, and 3D nnUnet (Isensee et al 2021); 2) previous prostate tumor segmentation approaches, including ProCDet (Qian, Zhang, and Wang 2021), MFSL-Net (Zhang et al 2021) and CSAD (Zhang et al 2022a); and 3) other multi-modal and crossslice interaction techniques, including F2Net (Yang et al 2023), ACMINet (Zhuang et al 2022), and CAT-Net (Hung et al 2022). For methods not designed for multi-modalities, we adapt them by concatenating three modalities as multichannel inputs while maintaining the original network structure.…”

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 99%

NaMa: Neighbor-Aware Multi-Modal Adaptive Learning for Prostate Tumor Segmentation on Anisotropic MR Images

Meng,

Zhang,

Huang

et al. 2024

AAAI

View full text Add to dashboard Cite

Accurate segmentation of prostate tumors from multi-modal magnetic resonance (MR) images is crucial for diagnosis and treatment of prostate cancer. However, the robustness of existing segmentation methods is limited, mainly because these methods 1) fail to adaptively assess subject-specific information of each MR modality for accurate tumor delineation, and 2) lack effective utilization of inter-slice information across thick slices in MR images to segment tumor as a whole 3D volume. In this work, we propose a two-stage neighbor-aware multi-modal adaptive learning network (NaMa) for accurate prostate tumor segmentation from multi-modal anisotropic MR images. In particular, in the first stage, we apply subject-specific multi-modal fusion in each slice by developing a novel modality-informativeness adaptive learning (MIAL) module for selecting and adaptively fusing informative representation of each modality based on inter-modality correlations. In the second stage, we exploit inter-slice feature correlations to derive volumetric tumor segmentation. Specifically, we first use a Unet variant with sequence layers to coarsely capture slice relationship at a global scale, and further generate an activation map for each slice. Then, we introduce an activation mapping guidance (AMG) module to refine slice-wise representation (via information from adjacent slices) for consistent tumor segmentation across neighboring slices. Besides, during the network training, we further apply a random mask strategy to each MR modality to improve feature representation efficiency. Experiments on both in-house and public (PICAI) multi-modal prostate tumor datasets show that our proposed NaMa performs better than state-of-the-art methods.

show abstract