用于低剂量CT图像去噪的多级双树复小波网络  

Multi-Level Dual-Tree Complex Wavelet Network for Low-Dose CT Image Denoising

作　　者：张鲁 田春伟 宋焕生[1,4] 刘侍刚[5] ZHANG Lu;TIAN Chunwei;SONG Huansheng;LIU Shigang(Educational Technology and Network Center,Chang′an University,Xi′an 710064,Shaanxi,China;School of Software,Northwestern Polytechnical University,Xi′an 710072,Shaanxi,China;Shenzhen Research Institute,Northwestern Polytechnical University,Shenzhen 518057,Guangdong,China;School of Information Engineering,Chang′an University,Xi′an 710064,Shaanxi,China;School of Computer Science,Shaanxi Normal University,Xi′an 710119,Shaanxi,China)

机构地区：[1]长安大学教育技术与网络中心,陕西西安710064 [2]西北工业大学软件学院,陕西西安710072 [3]西北工业大学深圳研究院,广东深圳518057 [4]长安大学信息工程学院,陕西西安710064 [5]陕西师范大学计算机科学学院,陕西西安710119

出　　处：《计算机工程》2024年第9期266-275,共10页Computer Engineering

基　　金：广东省基础与应用基础研究基金(2021A1515110079);深圳市科技创新委员会项目(JSGG20220831105002004);中国博士后科学基金(2022TQ0259,2022M722599)。

摘　　要：基于卷积神经网络(CNN)的图像去噪方法能有效去除低剂量计算机断层扫描(CT)图像伴随的伪影和噪声,从而确保CT设备输出高质量图像同时降低辐射,这对患者健康和医学诊断具有重要意义。为了进一步提高低剂量CT图像的质量,提出一种小波域去噪网络MDTNet。首先,基于双树复小波变换(DTCWT)构造多级编解码去噪网络,在多个尺度上提取特征以保留更多高频细节;然后,利用扩展的像素重排技术替代卷积上下采样,实现多级输入和特征融合,从而降低计算复杂度;最后,通过大量训练找到最佳的去噪模型,即二级MDTNet配合LeGall滤波器和Qshift_b滤波器,并选择较大尺寸的CT图像作为训练数据。使用AAPM数据集评估MDTNet的性能,实验结果表明,MDTNet能有效去除条纹状伪影和噪声,在定量和定性评估中性能均优于同类型去噪方法。与FWDNet相比,对于1 mm的切片,MDTNet的平均峰值信噪比(PSNR)和结构相似性指数(SSIM)分别提高了0.0887 dB和0.0024;对于3 mm的切片,分别提升了0.1443 dB和0.003。对于单张512×512像素的低剂量CT图像去噪,MDTNet在GPU上仅需0.193 s。MDTNet在保持高效率的同时保留了更多的高频细节,能够为低剂量CT图像去噪提供一种新的框架。Based on the Convolutional Neural Network(CNN),image denoising methods can effectively remove the artifacts and noise associated with low-dose Computed Tomography(CT),thereby ensuring high-quality output while minimizing radiation exposure.This information is of great significance for patient health and medical diagnosis.This study proposes a novel denoising network called MDTNet to enhance the quality of low-dose CT images.In this approach,multilevel encoder-decoder denoising networks are constructed using a Dual-Tree Complex Wavelet Transform(DTCWT),which enables the preservation of high-frequency details.Moreover,a pixel shuffle was employed to facilitate multi-level input and feature fusion,resulting in significantly reduced computation and memory complexity.In addition,by training a set of residual mappings in the wavelet domain,optimal denoising performance was achieved using a two-level MDTNet with LeGall and Qshift_b filters.The effectiveness of the MDTNet was evaluated on the 2016 NIH-AAPM-Mayo Clinic low-dose CT grand challenge dataset.The experimental results demonstrate that MDTNet outperforms state-of-the-art denoising methods in quantitative and qualitative evaluations.Specifically,compared with the FWDNet,on 1 mm slices,MDTNet improved the average Peak Signal-to-Noise Ratio(PSNR)and Structural Similarity Index Measure(SSIM)by 0.0887 dB and 0.0024,respectively;on a 3 mm slice,the increase was 0.1443 dB and 0.003,respectively.Moreover,MDTNet processed 512×512 low-dose CT images on a Graphical Processing Unit(GPU)in 0.193 s.Preserving high-frequency details while maintaining efficiency,MDTNet presents an innovative framework for denoising low-dose CT images.

关 键 词：低剂量CT图像 图像去噪 卷积神经网络 双树复小波变换 像素重排 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]