Frequency division denoising algorithm based on VIF adaptive 2D-VMD ultrasound image

Abstract

Ultrasound imaging has developed into an indispensable imaging technology in medical diagnosis and treatment applications due to its unique advantages, such as safety, affordability, and convenience. With the development of data information acquisition technology, ultrasound imaging is increasingly susceptible to speckle noise, which leads to defects, such as low resolution, poor contrast, spots, and shadows, which affect the accuracy of physician analysis and diagnosis. To solve this problem, we proposed a frequency division denoising algorithm combining transform domain and spatial domain. First, the ultrasound image was decomposed into a series of sub-modal images using 2D variational mode decomposition (2D-VMD), and adaptively determined 2D-VMD parameter K value based on visual information fidelity (VIF) criterion. Then, an anisotropic diffusion filter was used to denoise low-frequency sub-modal images, and a 3D block matching algorithm (BM3D) was used to reduce noise for high-frequency images with high noise. Finally, each sub-modal image was reconstructed after processing to obtain the denoised ultrasound image. In the comparative experiments of synthetic, simulation, and real images, the performance of this method was quantitatively evaluated. Various results show that the ability of this algorithm in denoising and maintaining structural details is significantly better than that of other algorithms.

Evaluation of left ventricular diastolic function of patients with coronary heart disease by ultrasound images on bilateral filtering image noise reduction algorithm combined with electrocardiogram

“… 9 In recent years, the methods used for ultrasonic image denoising can be classified into adaptive denoising, anisotropic diffusion denoising, non-local mean denoising, multi-scale denoising, and hybrid denoising. 10 , 11 However, the denoising algorithm using only wavelet transform is not effective in removing speckle noise in medical ultrasound images. 12 For bilateral filters, it has a strong denoising effect while also maintaining image edge details when dealing with image noise.…”

Medical image denoising based on 2D adaptive compact variational mode decomposition

“…In [12], the decomposition modal number K of the ultrasound images is set as 3, which is consistent with the results of the algorithm proposed in this paper. However, method [12] not adaptively optimize the penalty factor α , nor does it introduce a binary support function to increase the space compactness, which is the innovation and superiority of the algorithm proposed in this paper. From the analysis of the results, the image denoised by 2D-ACVMD is clearer and the details are more complete, and is improved by 1.3dB in PSNR and 0.02 in SSIM.…”

“…PSNR SSIM SRAD [10] 15.4716 0.3906 BM3D [11] 15.5129 0.4132 2D-VMD-VIF [12] 16.1197 0.4464 2D-ACVMD 17.4321 0.4653…”

Inverse radon transform with deep learning: an application in cardiac motion correction

“…Recently, the famous Alphafold-2 also used attention mechanism to predict protein structures from input amino acid sequences (Jumper et al 2021). It was designed for learning the long-range association between data (Bahdanau et al 2014, Vaswani et al 2017, Wang et al 2018, similar to how the semantics of a word in a sentence may be determined by its context. Some recent works (Zhang et al 2019) show that the attention component can be embedded into a CNN to enhance the image processing quality, as we have presented here.…”

“…Our approach (figure 5) is based on the CED architecture, originally proposed in (Häggström et al 2019), enhanced with the self-attention mechanism, which we call Convolutional Encoder-Decoder with Attention or CEDA (Wang et al 2018, Zhang et al 2019. The conventional CED architecture contains two parts: the encoder and the decoder.…”