Limits...
Vascular tree segmentation in medical images using Hessian-based multiscale filtering and level set method.

Jin J, Yang L, Zhang X, Ding M - Comput Math Methods Med (2013)

Bottom Line: Then Hessian-based multiscale filtering is used to enhance vascular structures by combining Hessian matrix with Gaussian convolution to tune the filtering response to the specific scales.Our approach was tested on synthetic images with vascular-like structures and 2D slices extracted from real 3D abdomen magnetic resonance angiography (MRA) images along the coronal plane.The segmentation rates for synthetic images are above 95%.

View Article: PubMed Central - PubMed

Affiliation: Department of Biomedical Engineering, School of Life Science and Technology, Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China, Huazhong University of Science and Technology, Wuhan 430074, China.

ABSTRACT
Vascular segmentation plays an important role in medical image analysis. A novel technique for the automatic extraction of vascular trees from 2D medical images is presented, which combines Hessian-based multiscale filtering and a modified level set method. In the proposed algorithm, the morphological top-hat transformation is firstly adopted to attenuate background. Then Hessian-based multiscale filtering is used to enhance vascular structures by combining Hessian matrix with Gaussian convolution to tune the filtering response to the specific scales. Because Gaussian convolution tends to blur vessel boundaries, which makes scale selection inaccurate, an improved level set method is finally proposed to extract vascular structures by introducing an external constrained term related to the standard deviation of Gaussian function into the traditional level set. Our approach was tested on synthetic images with vascular-like structures and 2D slices extracted from real 3D abdomen magnetic resonance angiography (MRA) images along the coronal plane. The segmentation rates for synthetic images are above 95%. The results for MRA images demonstrate that the proposed method can extract most of the vascular structures successfully and accurately in visualization. Therefore, the proposed method is effective for the vascular tree extraction in medical images.

Show MeSH

Related in: MedlinePlus

The segmentation results of the above three methods. (a) Original image. (b) Hessian-based multiscale filtering. (c) Hessian-based multiscale filtering combined with Chan-Vese model. (d) The proposed method.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3852584&req=5

fig6: The segmentation results of the above three methods. (a) Original image. (b) Hessian-based multiscale filtering. (c) Hessian-based multiscale filtering combined with Chan-Vese model. (d) The proposed method.

Mentions: In this section, we compared the segmentation results of the proposed method with those of other two vessel segmentation techniques, Hessian-based multiscale filtering [5] and Hessian-based multiscale filtering combined with Chan-Vese model [24], on the synthetic image. The synthetic image of size 512 × 512 used in this part is given in Figure 6(a). Figure 6(b) shows that the Hessian-based multiscale filtering can locate vessel structures accurately but with inaccurate scales, which means that it is suitable to generate the initial contour. Hessian-based multiscale filtering combined with Chan-Vese model can converge to the boundary of the vascular structure exactly, but it will leak into neighboring nonvascular structures where the contrast is low as shown in Figure 6(c). The result of the proposed method presented in Figure 6(d) is of high accuracy and completely unaffected by nonvascular structures because of the introduction of a new class of external constrained term Fσ to penalize the evolution distance of the contour. The segmentation rate of Hessian-based multiscale filtering, Hessian-based multiscale filtering combined with Chan-Vese model, and the proposed method are 88.71%, 92.32%, and 98.14%, respectively.


Vascular tree segmentation in medical images using Hessian-based multiscale filtering and level set method.

Jin J, Yang L, Zhang X, Ding M - Comput Math Methods Med (2013)

The segmentation results of the above three methods. (a) Original image. (b) Hessian-based multiscale filtering. (c) Hessian-based multiscale filtering combined with Chan-Vese model. (d) The proposed method.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3852584&req=5

fig6: The segmentation results of the above three methods. (a) Original image. (b) Hessian-based multiscale filtering. (c) Hessian-based multiscale filtering combined with Chan-Vese model. (d) The proposed method.
Mentions: In this section, we compared the segmentation results of the proposed method with those of other two vessel segmentation techniques, Hessian-based multiscale filtering [5] and Hessian-based multiscale filtering combined with Chan-Vese model [24], on the synthetic image. The synthetic image of size 512 × 512 used in this part is given in Figure 6(a). Figure 6(b) shows that the Hessian-based multiscale filtering can locate vessel structures accurately but with inaccurate scales, which means that it is suitable to generate the initial contour. Hessian-based multiscale filtering combined with Chan-Vese model can converge to the boundary of the vascular structure exactly, but it will leak into neighboring nonvascular structures where the contrast is low as shown in Figure 6(c). The result of the proposed method presented in Figure 6(d) is of high accuracy and completely unaffected by nonvascular structures because of the introduction of a new class of external constrained term Fσ to penalize the evolution distance of the contour. The segmentation rate of Hessian-based multiscale filtering, Hessian-based multiscale filtering combined with Chan-Vese model, and the proposed method are 88.71%, 92.32%, and 98.14%, respectively.

Bottom Line: Then Hessian-based multiscale filtering is used to enhance vascular structures by combining Hessian matrix with Gaussian convolution to tune the filtering response to the specific scales.Our approach was tested on synthetic images with vascular-like structures and 2D slices extracted from real 3D abdomen magnetic resonance angiography (MRA) images along the coronal plane.The segmentation rates for synthetic images are above 95%.

View Article: PubMed Central - PubMed

Affiliation: Department of Biomedical Engineering, School of Life Science and Technology, Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China, Huazhong University of Science and Technology, Wuhan 430074, China.

ABSTRACT
Vascular segmentation plays an important role in medical image analysis. A novel technique for the automatic extraction of vascular trees from 2D medical images is presented, which combines Hessian-based multiscale filtering and a modified level set method. In the proposed algorithm, the morphological top-hat transformation is firstly adopted to attenuate background. Then Hessian-based multiscale filtering is used to enhance vascular structures by combining Hessian matrix with Gaussian convolution to tune the filtering response to the specific scales. Because Gaussian convolution tends to blur vessel boundaries, which makes scale selection inaccurate, an improved level set method is finally proposed to extract vascular structures by introducing an external constrained term related to the standard deviation of Gaussian function into the traditional level set. Our approach was tested on synthetic images with vascular-like structures and 2D slices extracted from real 3D abdomen magnetic resonance angiography (MRA) images along the coronal plane. The segmentation rates for synthetic images are above 95%. The results for MRA images demonstrate that the proposed method can extract most of the vascular structures successfully and accurately in visualization. Therefore, the proposed method is effective for the vascular tree extraction in medical images.

Show MeSH
Related in: MedlinePlus