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

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.

Download Full-text

Weighted area constraints-based breast lesion segmentation in ultrasound image analysis

Inverse Problems and Imaging ◽

10.3934/ipi.2021057 ◽

2021 ◽

Vol 0 (0) ◽

pp. 0

Author(s):

Qianting Ma ◽

Tieyong Zeng ◽

Dexing Kong ◽

Jianwei Zhang

Keyword(s):

Level Set ◽

Domain Knowledge ◽

Ultrasound Image ◽

Speckle Noise ◽

Breast Ultrasound ◽

Nonlinear Constraint ◽

Low Contrast ◽

Anisotropic Diffusion Filter ◽

Initial Segmentation ◽

Speed Up

<p style='text-indent:20px;'>Breast ultrasound segmentation is a challenging task in practice due to speckle noise, low contrast and blurry boundaries. Although numerous methods have been developed to solve this problem, most of them can not produce a satisfying result due to uncertainty of the segmented region without specialized domain knowledge. In this paper, we propose a novel breast ultrasound image segmentation method that incorporates weighted area constraints using level set representations. Specifically, we first use speckle reducing anisotropic diffusion filter to suppress speckle noise, and apply the Grabcut on them to provide an initial segmentation result. In order to refine the resulting image mask, we propose a weighted area constraints-based level set formulation (WACLSF) to extract a more accurate tumor boundary. The major contribution of this paper is the introduction of a simple nonlinear constraint for the regularization of probability scores from a classifier, which can speed up the motion of zero level set to move to a desired boundary. Comparisons with other state-of-the-art methods, such as FCN-AlexNet and U-Net, show the advantages of our proposed WACLSF-based strategy in terms of visual view and accuracy.</p>

Download Full-text

Weighted Image Averaging Based Anisotropic Diffusion Denoising Method for Ultrasound Thyroid Image

Journal of Medical Imaging and Health Informatics ◽

10.1166/jmihi.2020.2902 ◽

2020 ◽

Vol 10 (2) ◽

pp. 380-390

Author(s):

Haiyue Zhang ◽

Daoyun Xu ◽

Yongbin Qin

Keyword(s):

Thyroid Disease ◽

Anisotropic Diffusion ◽

Averaging Method ◽

Ultrasound Image ◽

Speckle Noise ◽

Denoising Method ◽

Weighted Image ◽

Diffusion Filter ◽

Anisotropic Diffusion Filter ◽

Image Averaging

Thyroid disease is a frequent occurrence in clinical practice and the computerized analysis of ultrasonography has been becoming the most prospective tool for thyroid disease automatic diagnosis. However, the accuracy of vision-based diagnostic analysis is often reduced because the quality of ultrasound image is easily corrupted by the speckle noise. Thus, noise suppression is imperative and significant for the thyroid ultrasonography image preprocessing to increase the reliability of subsequent analysis. In this paper, we propose a novel weighted image averaging method based on anisotropic diffusion filters combination to remove speckle noise and enhance the details of the image at the same time. The method first denoises the image separately by two filters with different performances. The speckle reducing anisotropic diffusion filter can enhance the details of the image, and the anisotropic diffusion filter can better suppress the speckle noise in the image. In order to integrate the advantages of the two filters and reduce the mutual interference meanwhile, an adaptive weighted image averaging method is further proposed to combine the pixels of the two denoised images. The experimental results indicate that the proposed method can achieve promising performance on the template images with various noise levels by considering PSNR and SSIM. What's more, it is not only superior to other methods in automatic segmentation, but also can obtain better visual effect for thyroid images.

Download Full-text

Research on ultrasound image of interventional catheterization processing method based on wavelet transform and fuzzy theory

Filomat ◽

10.2298/fil2015187l ◽

2020 ◽

Vol 34 (15) ◽

pp. 5187-5194

Author(s):

Chenyang Liang ◽

Ning He

Keyword(s):

Wavelet Transform ◽

Fuzzy Theory ◽

Ultrasound Image ◽

Low Frequency ◽

Speckle Noise ◽

Ultrasound Images ◽

Wavelet Coefficients ◽

Image Processing Algorithm ◽

Exponential Transformation ◽

Interventional Catheterization

Interventional catheterization can help patients to accurately assess the condition, early diagnosis and intervention. Confirming the location of catheter by ultrasound has the advantages of real-time imaging, non-invasive, radiative, fast and convenient. Due to speckle noise and similar acoustic impedance, ultrasound images are not clear. In this paper an ultrasonic image processing algorithm based on wavelet transform and fuzzy theory is proposed. First, logarithmic transformation of ultrasound images is used to convert multiplicative noise into additive noise. Then the wavelet coefficients of the image are obtained by multiscale wavelet transform. The high frequency wavelet coefficients of the image are denoised by thresholding, and the low-frequency wavelet coefficients of the image are processed by fuzzy enhancement. Finally, the processed image is obtained through wavelet reconstruction and exponential transformation. Experiments show that this proposed method can effectively improve the visual effect of images.

Download Full-text

Fuzzy Based Adaptive Anisotropic Diffusion Despeckling Filter

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.b7100.129219 ◽

2019 ◽

Vol 9 (2) ◽

pp. 2519-2522

Keyword(s):

Anisotropic Diffusion ◽

Multiplicative Noise ◽

Ultrasound Image ◽

Visual Quality ◽

Speckle Noise ◽

Structure And Function ◽

Diffusion Filter ◽

Anisotropic Diffusion Filter ◽

And Function

Echocardiography is an ultrasound of the heart for the assessment of cardiac structure and function. Images obtained from the ultrasound suffer from a speckle Noise. Speckle is an inherent multiplicative noise which affects the visual perception by discriminating the fine details in the echocardiography. Speckle removal is important step to improve the visual quality of the echocardiography for better diagnosis. Anisotropic diffusion is one of the popular techniques to despeckle the ultrasound image in recent times. In this paper, we propose a fuzzy based adaptive anisotropic diffusion despeckling filter to despeckle echocardiography ultrasound image. The results show that fuzzy diffusion when combined with adaptive anisotropic diffusion filter gives better performance compared with existing despeckling filters

Download Full-text

A Proposed Approach to De-speckle an Ultrasound Image using Anisotropic Filter by Generating Diffusion Tensor

International Journal of Current Engineering and Technology ◽

10.14741/ijcet/v.8.3.29 ◽

2018 ◽

Vol 8 (3) ◽

Author(s):

Poonam Chauhan ◽

Vikas Kaushik

Keyword(s):

Ultrasound Imaging ◽

Diffusion Tensor ◽

Signal To Noise Ratio ◽

Ultrasound Image ◽

Similarity Index ◽

Speckle Noise ◽

The Body ◽

Invasive Technique ◽

Signal To Noise ◽

Noise Ratio

Ultrasound imaging is a technique that is used to diagnose the diseases in medical field using radiology. US (ultrasound) imaging is a non -invasive technique and used for imaging of internal structure of the body without any kind of penetration which helps to identify the diseases that have probability and tissues. Many kinds of noises present in US images but the presence of speckle noise is a big challenge since last few years in biomedical field. Sometimes speckle noise becomes the part of information and vice-versa. So it becomes hard to find the disease for doctors. There are many de-speckled filters available for de-noising. This paper gives a proposed approach to de-speckled the US image using anisotropic diffusion filter by calculating the different numerical values like SSIM (structural similarity index), SNR (signal to noise ratio), MSE (mean square error), PSNR (peak signal to noise ratio), which results in coherence enhancement The proposed technique provides better and improved edge and coherence enhancement in ultrasound image data.

Download Full-text

A Review on Various Ultrasound Image Despeckling Techniques

10.21467/proceedings.114.16 ◽

2021 ◽

Author(s):

Abhinav Panwar ◽

Ramesh K. Sunkaria ◽

Anterpreet Bedi

Keyword(s):

Ultrasound Imaging ◽

Ultrasound Image ◽

Speckle Noise ◽

Imaging Modalities ◽

Disease Detection ◽

Ultrasound Images ◽

Non Invasive ◽

Important Phenomenon ◽

Image Despeckling

Ultrasound imaging is considered as one of the most widely used imaging modalities owing to its simple and non-invasive nature. However, ultrasound images are usually manifested with speckle noise, that acts as a hindrance in carrying out any further analysis or disease detection by the radiologists. Despeckling of these images is thus a very important phenomenon to carry out further studies by medical experts. It is of utmost importance that ultrasound images be despeckled, keeping in consideration that no information is lost from the images. This paper covers various despeckling techniques that have been designed for ultrasound images, making sure that no information is lost.

Download Full-text

A Random Sampling Model of Visual Information Acquisition

PsycEXTRA Dataset ◽

10.1037/e665412011-558 ◽

1992 ◽

Author(s):

Geoffrey Loftus ◽

Thomas Busey ◽

John Senders

Keyword(s):

Information Acquisition ◽

Random Sampling ◽

Visual Information

Download Full-text

Reverse Scan Conversion and Efficient Deep Learning Network Architecture for Ultrasound Imaging on a Mobile Device

Sensors ◽

10.3390/s21082629 ◽

2021 ◽

Vol 21 (8) ◽

pp. 2629

Author(s):

Kunkyu Lee ◽

Min Kim ◽

Changhyun Lim ◽

Tai-Kyong Song

Keyword(s):

Deep Learning ◽

Ultrasound Imaging ◽

Mobile Device ◽

Network Architecture ◽

Point Of Care ◽

Ultrasound Image ◽

Training Dataset ◽

Point Of Care Ultrasound ◽

Reverse Scan ◽

Scan Conversion

Point-of-care ultrasound (POCUS), realized by recent developments in portable ultrasound imaging systems for prompt diagnosis and treatment, has become a major tool in accidents or emergencies. Concomitantly, the number of untrained/unskilled staff not familiar with the operation of the ultrasound system for diagnosis is increasing. By providing an imaging guide to assist clinical decisions and support diagnosis, the risk brought by inexperienced users can be managed. Recently, deep learning has been employed to guide users in ultrasound scanning and diagnosis. However, in a cloud-based ultrasonic artificial intelligence system, the use of POCUS is limited due to information security, network integrity, and significant energy consumption. To address this, we propose (1) a structure that simultaneously provides ultrasound imaging and a mobile device-based ultrasound image guide using deep learning, and (2) a reverse scan conversion (RSC) method for building an ultrasound training dataset to increase the accuracy of the deep learning model. Experimental results show that the proposed structure can achieve ultrasound imaging and deep learning simultaneously at a maximum rate of 42.9 frames per second, and that the RSC method improves the image classification accuracy by more than 3%.

Download Full-text