Chongqing University of Posts and Telecommunications, Chongqing 400065, China.
Chongqing University of Posts and Telecommunications, Chongqing 400065, China.
Comput Methods Programs Biomed. 2021 Jun;205:106110. doi: 10.1016/j.cmpb.2021.106110. Epub 2021 Apr 14.
For medical image segmentation, deep learning-based methods have achieved state-of-the-art performance. However, the powerful spectral representation in the field of image processing is rarely considered in these models.
In this work, we propose to introduce frequency representation into convolution neural networks (CNNs) and design a novel model, tKFC-Net, to combine powerful feature representation in both frequency and spatial domains. Through the Fast Fourier Transform (FFT) operation, frequency representation is employed on pooling, upsampling, and convolution without any adjustments to the network architecture. Furthermore, we replace original convolution with twin-Kernel Fourier Convolution (t-KFC), a new designed convolution layer, to specify the convolution kernels for particular functions and extract features from different frequency components.
We experimentally show that our method has an edge over other models in the task of medical image segmentation. Evaluated on four datasets-skin lesion segmentation (ISIC 2018), retinal blood vessel segmentation (DRIVE), lung segmentation (COVID-19-CT-Seg), and brain tumor segmentation (BraTS 2019), the proposed model achieves outstanding results: the metric F1-Score is 0.878 for ISIC 2018, 0.8185 for DRIVE, 0.9830 for COVID-19-CT-Seg, and 0.8457 for BraTS 2019.
The introduction of spectral representation retains spectral features which result in more accurate segmentation. The proposed method is orthogonal to other topology improvement methods and very convenient to be combined.
对于医学图像分割，基于深度学习的方法已经达到了最新的性能水平。然而，图像处理领域强大的频谱表示在这些模型中很少被考虑。
在这项工作中，我们提出将频率表示引入卷积神经网络（CNNs）中，并设计了一个新的模型 tKFC-Net，以结合频域和空域中的强大特征表示。通过快速傅里叶变换（FFT）操作，在池化、上采样和卷积中使用频率表示，而不对网络架构进行任何调整。此外，我们用双核傅里叶卷积（t-KFC）替换原始卷积，这是一种新设计的卷积层，用于指定特定函数的卷积核，并从不同的频率分量中提取特征。
我们通过实验表明，我们的方法在医学图像分割任务中优于其他模型。在四个数据集（皮肤病变分割（ISIC 2018）、视网膜血管分割（DRIVE）、肺部分割（COVID-19-CT-Seg）和脑肿瘤分割（BraTS 2019））上进行评估，所提出的模型取得了出色的结果：ISIC 2018 的 F1-Score 指标为 0.878，DRIVE 为 0.8185，COVID-19-CT-Seg 为 0.9830，BraTS 2019 为 0.8457。
引入频谱表示保留了光谱特征，从而实现了更精确的分割。所提出的方法与其他拓扑改进方法正交，并且非常便于组合。
