B-type ultrasound images have important applications in medical diagnosis. However, the widely spread intensity inhomogeneity, low-scale contrast, constructed defect, noise and blurred edges all make it difficult to implement automatic segmentation of lesion in the images. Based on region level set method, a subordinate degree region level set model was proposed, in which subordinate degree probability of each pixel was defined to reflect the pixel subjection grade to target and background respectively. Pixels were classified to either target or background by calculation of their subordinate degree probabilities, and edge contour was obtained by region level set iterations. In this paper, lesion segmentation is regarded as local segmentation of specific area, and the calculation is restrained to the local sphere abide by the contour, which greatly reduce the calculation complexity. Experiments on B-type ultrasound images showed improved results of the proposed method compared to those of some popular level set methods.
Algorithms ; Diagnostic Imaging ; Humans ; Models, Theoretical ; Probability ; Signal Processing, Computer-Assisted ; Ultrasonography

OBJECTIVETo modify the level set method for precise and fast segmentation of B-type ultrasound image lesions.

METHODSBased on the best of region level set method, entropy in the information theory was introduced into image processing to define a dynamic weighting factor that responded to the gradient change of the local gray levels to evaluate the dynamic degree of driven force on each pixel on the contour to the target and background areas. The dynamic weighting factors were reconciled into the regional level set model and led the contour to deform and move during the iterations. As lesion segmentation was classified as local segmentation of a specific area, the calculation was restrained to the local sphere abide by the contour, which greatly reduced the calculation complex.

RESULTSExperiments on B-type ultrasound images showed improved results of the proposed method with a better accuracy and less time consumption compared with several current level set methods.

CONCLUSIONLevel set method reconciled with dynamic weighting factor allows a better evaluation of the lesion contour pixels, and the local calculation strategy results in an enhanced segmentation efficiency.

Algorithms ; Image Processing, Computer-Assisted ; Models, Theoretical ; Ultrasonography

Objective To modify the level set method for precise and fast segmentation of B-type ultrasound image lesions. Methods Based on the best of region level set method, entropy in the information theory was introduced into image processing to define a dynamic weighting factor that responded to the gradient change of the local gray levels to evaluate the dynamic degree of driven force on each pixel on the contour to the target and background areas. The dynamic weighting factors were reconciled into the regional level set model and led the contour to deform and move during the iterations. As lesion segmentation was classified as local segmentation of a specific area, the calculation was restrained to the local sphere abide by the contour, which greatly reduced the calculation complex. Results Experiments on B-type ultrasound images showed improved results of the proposed method with a better accuracy and less time consumption compared with several current level set methods. Conclusion Level set method reconciled with dynamic weighting factor allows a better evaluation of the lesion contour pixels, and the local calculation strategy results in an enhanced segmentation efficiency.

Objective To modify the level set method for precise and fast segmentation of B-type ultrasound image lesions. Methods Based on the best of region level set method, entropy in the information theory was introduced into image processing to define a dynamic weighting factor that responded to the gradient change of the local gray levels to evaluate the dynamic degree of driven force on each pixel on the contour to the target and background areas. The dynamic weighting factors were reconciled into the regional level set model and led the contour to deform and move during the iterations. As lesion segmentation was classified as local segmentation of a specific area, the calculation was restrained to the local sphere abide by the contour, which greatly reduced the calculation complex. Results Experiments on B-type ultrasound images showed improved results of the proposed method with a better accuracy and less time consumption compared with several current level set methods. Conclusion Level set method reconciled with dynamic weighting factor allows a better evaluation of the lesion contour pixels, and the local calculation strategy results in an enhanced segmentation efficiency.