A New Variational Method for Bias Correction and Its Applications to Rodent Brain Extraction

Chang, Huibin; Huang, Weimin; Wu, Chunlin; Huang, Su; Guan, Cuntai; Sekar, S.; Bhakoo, Kishore; Duan, Yuping

doi:10.1109/tmi.2016.2636026

Cited by 29 publications

(24 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…FN is the total number of pixels incorrectly classified as non-brain tissue. Sensitivity represents the ability of brain extraction methods to correctly recognize brain tissue: (6) Specificity represents the ability of brain extraction methods to correctly recognize non-brain tissues. (7) Values of dice coefficient, sensitivity, and specificity range from 0 to 1.…”

Section: B Evaluation Metricsmentioning

confidence: 99%

“…Since the development of the brain extraction research, a lot of automatic methods have been proposed. These methods can be divided into the classic-based [1][2][3][4][5][6][7], the atlas-based [8][9], and the learning-based [10][11][12][13][14][15][16][17][18][19][20]. Some classic automatic methods are yet widely used because of their advantages in high calculation speed and batch processing of data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Brain Extraction From Brain MRI Images Based on Wasserstein GAN and O-Net

Jiang

Guo

Cheng³

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Brain extraction is an essential pre-processing step for neuroimaging analysis. It is difficult to achieve high-precision extraction from low-quality brain MRI images with artifacts and gray inconsistencies which often result in irregular hole regions in the extracted brain tissues. In addition, the U-Net based brain extraction methods trend to output over-smoothed brain boundary. To remove those irregular holes in the extracted mask, we proposed a new U-Net based model for brain extraction named O-Net. O-Net replaces the skip-connection path in the U-Net with a dual shortcut paths including an attention module to form an Oshaped network, which uses deep semantic information to highlight the target area while retaining more image details. O-Net effectively reduces the impact of intensity differences resulted from artifacts or gray inconsistencies in the brain MRI images on the extraction results. To obtain more precise brain boundary, we designed a new GAN based brain extraction method, which used above O-Net as the segmentation network. The discriminant network of the proposed GAN model adopts the residual structure to enhance the nonlinear expression ability of the network to balance the adversarial training of the two networks. To speed up the convergence of the proposed model, we added a segmentation loss to the adversarial loss to supervise the feature learning of the segmentation network. This method was compared with other popular brain extraction methods on two public datasets (IBSR18 and LPBA40). The mean dice similarity coefficients obtained by the proposed method were 97.26% and 98.29% on IBSR18 and LPBA40 respectively. The results of the proposed method are the best on the two public datasets. Experimental results prove that the proposed model can stably output high-precision brain tissue and reduce the influence of artifacts and gray inconsistencies.

show abstract

Section: B Evaluation Metricsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Brain Extraction From Brain MRI Images Based on Wasserstein GAN and O-Net

Jiang

Guo

Cheng³

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Unlike other methods, L0MS model gives an exact piecewise constant solution of r, which is a preliminary results for structural segmentation. Thus, the 0 regularized Retinex model has been reformulated for three-dimensional applications based on the highorder regularization in [9].…”

Section: Hotvl1 Model Recently Liang and Zhangmentioning

confidence: 99%

Retinex based on exponent-type total variation scheme

Liu¹,

Pang²,

Duan³

2018

Inverse Problems &Amp; Imaging

Self Cite

View full text Add to dashboard Cite

Retinex theory deals with compensation for illumination effects in images, which has a number of applications including Retinex illusion, medical image intensity inhomogeneity and color image shadow effect etc.. Such ill-posed problem has been studied by researchers for decades. However, most exiting methods paid little attention to the noises contained in the images and lost effectiveness when the noises increase. The main aim of this paper is to present a general Retinex model to effectively and robustly restore images degenerated by both illusion and noises. We propose a novel variational model by incorporating appropriate regularization technique for the reflectance component and illumination component accordingly. Although the proposed model is non-convex, we prove the existence of the minimizers theoretically. Furthermore, we design a fast and efficient alternating minimization algorithm for the proposed model, where all subproblems have the closed-form solutions. Applications of the algorithm to various gray images and color images with noises of different distributions yield promising results.

show abstract

“…However, this approach used only 2D https://doi.org/10.1016/j.phro.2017.11.003 Received 28 June 2017; Received in revised form 23 November 2017; Accepted 23 November 2017 T instead of the entire 3D image information to estimate the bias correction field which may reduce the correction performance. In contrast, Chang et al [15] proposed a higher-order variational model for bias correction for brain MR scans. Furthermore, Ivanovska et al [16] presented a level-set based approach for simultaneous intensity nonuniformity correction and segmentation of MR images.…”

Section: Introductionmentioning

confidence: 99%

Physical correction model for automatic correction of intensity non-uniformity in magnetic resonance imaging

Leger

Löck

Hietschold

et al. 2017

Physics and Imaging in Radiation Oncology

View full text Add to dashboard Cite

Background and purpose: Magnetic resonance imaging (MRI) plays an important role in the field of MR-guided radiotherapy or personalised radiation oncology. The application of quantitative image analyses like radiomics as well as automated tissue characterisation is frequently disturbed by the effect of intensity non-uniformity. We present a novel fully automated physical correction model (PCM) for the reduction of intensity non-uniformity. Materials and methods: The proposed algorithm is based on a 3D physically motivated correction model, which maximises the image information expressed by the Shannon entropy. The PCM was evaluated using the coefficient of variation (cv) on 176 MRI datasets of the human brain and abdomen acquired on 1.5 Tesla and 3 Tesla MR scanners. The resulting cv was compared to the cv of the original images and to the results of the established N4 algorithm. Results: The PCM algorithm significantly improved the image quality of all considered 1.5 and 3.0 Tesla MR scans compared to the original images (p < .01). Furthermore, the PCM outperformed or competed with the N4 algorithm in terms of image quality. Additionally, the PCM approach preserved the tissue signal of different tissue types due to smooth correction gradients. Conclusion: The proposed PCM algorithm led to a significantly improved image quality compared to the originally acquired images, suggesting that it is applicable to the correction of MRI data. Thus it may help to reduce intensity non-uniformity which is an important step for advanced image analysis.

show abstract

A New Variational Method for Bias Correction and Its Applications to Rodent Brain Extraction

Cited by 29 publications

References 54 publications

Brain Extraction From Brain MRI Images Based on Wasserstein GAN and O-Net

Brain Extraction From Brain MRI Images Based on Wasserstein GAN and O-Net

Retinex based on exponent-type total variation scheme

Physical correction model for automatic correction of intensity non-uniformity in magnetic resonance imaging

Contact Info

Product

Resources

About