Improved Segmentation of MRI Brain Images by Denoising and Contrast Enhancement

Sasirekha, N.; Kashwan, K. R.

doi:10.17485/ijst/2015/v8i22/73050

Cited by 12 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to improve the performance of median denoising algorithm, a variety of improved median denoising algorithms are proposed. As an extension of the median denoising algorithm, the statistical sorting denoising algorithm also shows good performance 12–14 . At the same time, some scholars combine nonlinear theory with mean algorithm and propose a variety of adaptive weighted mean denoising algorithm, so that the traditional linear denoising idea has been sublimated very well.…”

Section: Image Denoisingtechnologymentioning

confidence: 99%

Retracted: Complex image denoising framework with CNN‐wavelet under concurrency scenarios for informatics systems

Wang

Huang

Zhang

et al. 2018

Concurrency and Computation

View full text Add to dashboard Cite

Image denoising based on convolutional neural networks and wavelet transform is a novel approach for the applications. In the image acquisition process, images are often contaminated by noise, which affects the image quality; therefore, it is necessary to eliminate noise before analyzing and using images. Wavelet analysis is a local analysis method with multi-resolution characteristics, which is developed on the basis of short-time Fourier transform. It can be used for the multi-scale analysis of signals by means of expansion, translation and other operations, and extracting effective information from signals, which is a powerful tool for analyzing non-stationary signals. Wavelet has good time-frequency local characteristics, low entropy, and decorrelation. In this paper, we propose MRI image denoising framework based on convolutional neural networks and wavelet transform, and the experiment results show that the proposed method can keep the edge and curvature structure better while denoising. Compared with the other novel methodologies, the proposed algorithm can provide the higher robustness. In the future research, we will try the implementations of the methodologies.

show abstract

Section: Image Denoisingtechnologymentioning

confidence: 99%

Retracted: Complex image denoising framework with CNN‐wavelet under concurrency scenarios for informatics systems

Wang

Huang

Zhang

et al. 2018

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…The second application of our proposed framework is brain segmentation for MRI data. However, it is a difficult task owing to the artifacts and in-homogeneities introduced during the real image acquisition 21,22 . To this point, we propose a novel approach for brain segmentation, called translated multichannel segmentation (TMS).…”

mentioning

confidence: 99%

MRI Cross-Modality Image-to-Image Translation

Yang

Zhao

et al. 2020

Sci Rep

View full text Add to dashboard Cite

We present a cross-modality generation framework that learns to generate translated modalities from given modalities in MR images. Our proposed method performs Image Modality Translation (abbreviated as IMT) by means of a deep learning model that leverages conditional generative adversarial networks (cGANs). Our framework jointly exploits the low-level features (pixel-wise information) and high-level representations (e.g. brain tumors, brain structure like gray matter, etc.) between cross modalities which are important for resolving the challenging complexity in brain structures. Our framework can serve as an auxiliary method in medical use and has great application potential. Based on our proposed framework, we first propose a method for cross-modality registration by fusing the deformation fields to adopt the cross-modality information from translated modalities. Second, we propose an approach for MRI segmentation, translated multichannel segmentation (TMS), where given modalities, along with translated modalities, are segmented by fully convolutional networks (FCN) in a multichannel manner. Both of these two methods successfully adopt the crossmodality information to improve the performance without adding any extra data. Experiments demonstrate that our proposed framework advances the state-of-the-art on five brain MRI datasets. We also observe encouraging results in cross-modality registration and segmentation on some widely adopted brain datasets. Overall, our work can serve as an auxiliary method in medical use and be applied to various tasks in medical fields.

show abstract

“…There are various special filters available in image processing, and among them selection of optimal filter for the particular application is really a crucial task. In MRI, Rician noise [6] is present often, in order to remove the noise Gaussian filter is used here. Figure 3(b) shows Gaussian filtered image, where sharpness of image is got reduced, and noise became inefficient so that it will not have adverse effect on result [7].…”

Section: Methodology a Filter Imagementioning

confidence: 99%

An Efficacious Graphical User Interface Implementation for Automatic Classification of Brain Tumor from Magnetic Resonance Imaging Images Using Image Processing

Modi¹

2018

IJRASET

View full text Add to dashboard Cite

Biomedical image processing is an emerging field, and is very useful for automatic disease diagnosis. Brain tumor classification is performed using image processing toolbox of MATLAB over here. For tumor classification: Image Preprocessing, Segmentation and feature Analysis are essential steps, these steps are described in the paper. First of all, images were filtered by Gaussian filter, then tumor part i.e. regions having high intensity are enhanced using power low transform in order to neglect insufficient data from image. Then enhanced part was extracted and denoising was performed to get tumor region only at output. By the method we have got only tumor part having high intensity and all other part of brain considered as background. After tumor segmentation, feature analysis has been made by which we can get idea that how they differs into normal and benign or malignant tumors. Then classified based on feature analysis of tumor region, all the processes have been done over dataset of 45 Brain Tumor images. Here overall Accuracy of correct classification is got to be 97.78%, where normal brain is classified with 100%, benign with 95% and malignant with 100% accuracy. A Graphical User Interface (GUI) is developed in MATLAB to perform all the task and to test proposed work. In future, everything would be going to be computer assisted. This is a step toward automatic disease diagnosis using Computer Aided Design (CAD).

show abstract

Improved Segmentation of MRI Brain Images by Denoising and Contrast Enhancement

Cited by 12 publications

References 0 publications

Retracted: Complex image denoising framework with CNN‐wavelet under concurrency scenarios for informatics systems

Retracted: Complex image denoising framework with CNN‐wavelet under concurrency scenarios for informatics systems

MRI Cross-Modality Image-to-Image Translation

An Efficacious Graphical User Interface Implementation for Automatic Classification of Brain Tumor from Magnetic Resonance Imaging Images Using Image Processing

Contact Info

Product

Resources

About