An adaptive self-guided wavelet convolutional neural network with compound loss for low-dose CT denoising. (May 2022)

• Record Type: Journal Article
• Authors: Li, Saize; Li, Qing; Li, Runrui; Wu, Wei; Zhao, Juanjuan; Qiang, Yan; Tian, Yuling
• Is Part Of: Biomedical signal processing and control. Volume 75(2022)
• Publication Date: 2022-05
• Subjects: Low-dose CT -- Wavelet transform -- CNN -- Multi-scale and diversity -- Compound loss; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.103543
• Languages: English
• ISSNs: 1746-8094

Abstract

Graphical abstract: Excessive X-ray radiation during CT imaging can bring potential health risks. Reducing the radiation dose will increase the noise and artifacts of the reconstructed image and affect the doctor's early diagnosis. In order to improve the image quality of LDCT, we propose a deep learning method for LDCT image denoising. Highlights: An adaptive self-guided wavelet convolutional neural network (ASWCNN) for low-dose CT image denoising is proposed. In ASWCNN, a top-down self-guiding structure for changing the size of feature pictures is proposed by combining wavelet transforms and subpixel convolution. Secondly, the adjustable pyramid residual block (APRB) and cross-latitude mixed attention block (CMAB) are proposed, which adaptively extract the feature information of multiple scales and diversity of feature pictures and enhance the useful feature information and inhibit the useless information through the attention mechanism. Finally, a composite loss that combines pixel-level loss, structural perception loss, and gradient loss is proposed to convert low-dose CT images to normal-dose CT images as realistically as possible. Abstract: Computed Tomography (CT) is an imaging method widely used in clinical, industrial, and other applications. Furthermore, it is one of the common methods of modern clinical medical imaging diagnosis. However, excessive radiation doses in CT scans can cause harm to the human body. Reducing the radiation dose will cause serious degradation of the reconstructed image quality, produce speckle noise and streak artifacts, and affect the accuracy of clinical medical diagnosis. In order to obtain high-quality images while reducing CT radiation dose, we propose an adaptive self-guided wavelet convolutional neural network (ASWCNN) to convert low-dose CT (LDCT) images into normal-dose CT (NDCT) images as true as possible. In our ASWCNN, combining wavelet transform and sub-pixel convolution, a top-down self-guiding structure is proposed as the overall architecture of the network. An adjustable pyramid residual block (APRB) is proposed to self-adaption extract multi-scale and diversity information features as image resolution decreases along with the network. The adjacent scale information fusion block (ASIFB) is proposed to fuse the information features between adjacent scales step by step to improve the stability of network training. At the same time, we use the cross-latitude mixed attention block (CMAB) as the feature enhancement block of our network to enhance the fused feature information, enhance the effective information and suppress the useless information. In addition, we proposed the correction reconstruction block (CRB) to reduce the gap between the obtained denoised CT image and the given NDCT image. Finally, we take the compound loss combining the pixel-level loss, structure-perceived loss, and gradient loss as our loss function. Experimental results show that our proposed method can effectively retain the structure and texture information of CT images while removing noise and artifacts. At the same time, compared with other methods, better subjective visual evaluation and objective quantitative evaluation are obtained, and the network generalization ability is good.