用小波变换提升3D医学图像分割的效率与精度
WaveFormer: A 3D Transformer with Wavelet-Driven Feature Representation for Efficient Medical Image Segmentation
- 引入多尺度小波变换捕捉全局上下文与局部细节
- 参数量显著降低，推理速度更快，适配实际部署
- 生物启发设计，通用性强，适合多种医学影像任务
基于Transformer的架构在建模长程依赖方面推动了医学图像分析的发展，但在3D场景中常因内存开销大、难以捕捉细粒度局部特征而受限。我们提出WaveFormer，一种新型3D Transformer：i）利用特征的频域特性进行上下文表征；ii）受人类视觉识别系统自上而下机制启发，具有生物合理性。通过在多尺度上应用离散小波变换（DWT），WaveFormer同时保留全局上下文与高频细节，并以高效的小波摘要与重建替代复杂的上采样层，大幅减少参数量，对计算资源和训练时间受限的实际部署至关重要。该模型具备通用性，可灵活适配多种应用。在BraTS2023、FLARE2021和KiTS2023上的评估显示，其性能与顶尖方法相当，但计算复杂度显著更低。
原文摘要 · Abstract (English)
Transformer-based architectures have advanced medical image analysis by effectively modeling long-range dependencies, yet they often struggle in 3D settings due to substantial memory overhead and insufficient capture of fine-grained local features. We address these limitations with WaveFormer, a novel 3D-transformer that: i) leverages the fundamental frequency-domain properties of features for contextual representation, and ii) is inspired by the top-down mechanism of the human visual recognition system, making it a biologically motivated architecture. By employing discrete wavelet transformations (DWT) at multiple scales, WaveFormer preserves both global context and high-frequency details while replacing heavy upsampling layers with efficient wavelet-based summarization and reconstruction. This significantly reduces the number of parameters, which is critical for real-world deployment where computational resources and training times are constrained. Furthermore, the model is generic and easily adaptable to diverse applications. Evaluations on BraTS2023, FLARE2021, and KiTS2023 demonstrate performance on par with state-of-the-art methods while offering substantially lower computational complexity.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。