arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2312.04727cs.CV
收藏

E2ENet:用于精确高效3D医学图像分割的动态稀疏特征融合

E2ENet: Dynamic Sparse Feature Fusion for Accurate and Efficient 3D Medical Image Segmentation

University of Twente(特温特大学);University of Luxembourg(卢森堡大学);Eindhoven University of Technology(埃因霍温理工大学);University of Oxford(牛津大学);University of Surrey(萨里大学)

Boqian Wu, Qiao Xiao, Shiwei Liu, Lu Yin, Mykola Pechenizkiy, Decebal Constantin Mocanu, Maurice Van Keulen, Elena Mocanu

AI总结:

针对3D医学图像分割中模型规模与计算成本过高的问题,提出E2ENet模型,通过动态稀疏特征融合机制和3D卷积受限深度偏移设计,实现精度与效率的平衡,在多个数据集上表现更优且节省参数与计算量。

摘要:

深度神经网络因其出色的性能已成为3D医学图像分割的主流方法。然而,深度神经网络不断增长的模型规模和计算成本已成为在现实世界资源受限硬件上部署它们的主要障碍。为了同时提升性能和效率,我们提出了一种3D医学图像分割模型,名为高效到高效网络(E2ENet),该模型包含两种参数和计算高效的设计:i.动态稀疏特征融合(DSFF)机制:自适应学习融合多尺度信息特征,同时减少冗余;ii.3D卷积中的受限深度偏移:在保留3D空间信息的同时,保持模型和计算复杂度与基于2D的方法相当。我们在BTCV、AMOS-CT和脑肿瘤分割挑战赛上进行了大量实验,结果表明E2ENet在各种资源约束下始终比现有方法在精度和效率之间取得更优的平衡。在大规模挑战赛AMOS-CT上,E2ENet达到了与之前最佳性能方法相当的精度,同时在推理阶段节省了超过68%的参数数量和29%的FLOPs。我们的代码已开源:https://github.com/boqian333/E2ENet-Medical。

英文摘要:

Deep neural networks have evolved as the leading approach in 3D medical image segmentation due to their outstanding performance. However, the ever-increasing model size and computation cost of deep neural networks have become the primary barrier to deploying them on real-world resource-limited hardware. In pursuit of improving performance and efficiency, we propose a 3D medical image segmentation model, named Efficient to Efficient Network (E2ENet), incorporating two parametrically and computationally efficient designs. i. Dynamic sparse feature fusion (DSFF) mechanism: it adaptively learns to fuse informative multi-scale features while reducing redundancy. ii. Restricted depth-shift in 3D convolution: it leverages the 3D spatial information while keeping the model and computational complexity as 2D-based methods. We conduct extensive experiments on BTCV, AMOS-CT and Brain Tumor Segmentation Challenge, demonstrating that E2ENet consistently achieves a superior trade-off between accuracy and efficiency than prior arts across various resource constraints. E2ENet achieves comparable accuracy on the large-scale challenge AMOS-CT, while saving over 68\% parameter count and 29\% FLOPs in the inference phase, compared with the previous best-performing method. Our code has been made available at: https://github.com/boqian333/E2ENet-Medical.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑