Abstract
To improve the visual quality of noisy medical images acquired by low radiation dose imaging, medical image denoising is highly desirable for clinical disease diagnosis. In this paper, a geometric regularization method is proposed for medical image denoising. To the best of our knowledge, this is the first work aiming at reconstructing surface by minimizing the gradient error and approximation error of the surface to suppress the noise in medical images. The proposed denoising method consists of two stages: one is to output a basic estimate and the other is for the residual noise reduction. Specifically, the method first exploits a biquadratic polynomial surface to generate an initial estimate of the noise-free image. The surface is constructed by dividing its coefficients into two groups. With the reconstruction error constraint, one group is used to minimize the gradient of the surface, and the other is to minimize the approximation accuracy of the surface. Then the residual noise in the initial result is further reduced by using the singular value thresholding mechanism, which exploits the self-similarity of medical images and the intrinsic low-rank property. Unlike the traditional truncated singular value thresholding scheme, the proposed singular value thresholding is derived by optimizing an objective function with a constraint. Experimental results on a real clinical data set demonstrate the effectiveness of the proposed denoising method, especially in detail-preserving. Compared with several widely used denoising methods, our method can achieve a better performance in terms of both quantitative metrics and subjective visual quality.
Highlights
Medical imaging has been widely applied in clinical disease diagnosis with the advent of digital imaging technologies
The method first applies a biquadratic polynomial surface reconstruction algorithm to derive an initial denoised image, and the residual noise in the initial result is further reduced by a singular value shrinkage strategy
This paper presents a new method for medical image denoising that combines biquadratic polynomial surface construction and low-rank approximation techniques
Summary
Medical imaging has been widely applied in clinical disease diagnosis with the advent of digital imaging technologies. Where α is a sparse representation vector, and D denotes a representation dictionary that can be constructed by using the fixed wavelet/curvelet basis or adaptively learned by a greedy algorithm from the noisy images or given noise-free images [11]. Recent works have shown that low-rank priors are powerful models to reduce noise [12]–[14], in which an image is represented by a low-rank matrix. This model has been widely used to deal with various image restoration problems Many variants, such as weighted nuclear norm and low-rank and sparse combination priors, have been proposed to further improve its performance [16]–[18]. The method first applies a biquadratic polynomial surface reconstruction algorithm to derive an initial denoised image, and the residual noise in the initial result is further reduced by a singular value shrinkage strategy
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
