Low-dose CT noise reduction based on local total variation and improved wavelet residual CNN

Abstract

BACKGROUND: Low-dose computed tomography (LDCT) is an effective method for reducing radiation exposure. However, reducing radiation dose leads to considerable noise in the reconstructed image that can affect doctor’s judgment. OBJECTIVE: To solve this problem, this study proposes a local total variation and improved wavelet residual convolutional neural network (LTV-WRCNN) denoising model. METHODS: The model first introduces local total variation (LTV) to decompose the LDCT image into cartoon and texture image. Next, the texture image is filtered using the non-local mean (NLM). Then, the cartoon image is added to the filtered texture image to obtain the preprocessing image. Finally, the pre-processed image is fed into the improved wavelet residual neural network (WRCNN) to obtain an improved image. Additionally, we also introduce a compound loss in wavelet domain that combines mean squared error loss and directional regularization loss to separate the structural details from noise more thoroughly. RESULTS: Compared with state-of-the-art methods, the peak-signal-to-noise ratio (PSNR) value and the structure similarity (SSIM) value of the processed CT images using the new proposed model are 33.4229 dB and 0.9158. Study also shows that applying new model obtains better results visually and numerically, especially in terms of the preservation of structural details. CONCLUSIONS: The proposed new model is feasible and effective in improving the quality of LDCT images.

CT image denoising methods for image quality improvement and radiation dose reduction

“…The predominant DL models for CT denoising are GANs and CNNs. As shown in Figure 2a, out of 99 publications examined, 61 studies use the models based on CNN, 59–119 while 30 studies are based on GAN 120–149 . Additionally, two studies adopt Transformer‐based approaches 150,151 .…”

“…The majority of research focusing on the objective image quality evaluations of DL algorithms has consistently demonstrated remarkable noise reduction compared to FBP and IR at equivalent or lower radiation dose levels. 74,77,79,82,90,92,93,95,103,104,113,114,147 Additionally, the implementation of DL for metal artifact reduction demonstrates superior results when compared to IR. 62,86,119,121 CT image denoising approaches show promising potential, but are not widely accepted in routine clinical practice.…”

“…Supervised DL‐based methods have been shown to produce high‐quality denoised CT images, and they are effective at removing various types of noise patterns. Out of 99 studies reviewed, 64 studies 42,59–68,71,72,74,76,77,79,80,83,84,86,88–93,97–104 , 106–109,111–119,123–126,132,134,137–139,142,147,153,159,161–163 apply supervised training method. CNNs are a common type of supervised DL‐based model used for CT image denoising.…”

“…The potential application of DL‐based CT denoising models has been identified to enhance image quality, improve diagnostic accuracy and streamline radiological workflows. The majority of research focusing on the objective image quality evaluations of DL algorithms has consistently demonstrated remarkable noise reduction compared to FBP and IR at equivalent or lower radiation dose levels 74,77,79,82,90,92,93,95,103,104,113,114,147 . Additionally, the implementation of DL for metal artifact reduction demonstrates superior results when compared to IR 62,86,119,121 …”

Deep learning-based rectum segmentation on low-field prostate MRI to assist image-guided biopsy

“…2. Since clinical image quality is affected by multiple factors [7], including patient positioning relative to the MRI hardware, the applied MRI pulse sequences, and possible patient motion during image acquisition, we investigated and compared several image preprocessing methods to compensate for the possible negative effects of these variations in imaging acquisition. These methods include histogram equalization, contrasting stretching, and brightness enhancement [8].…”