A variational multiphase level set approach to simultaneous segmentation and bias correction

Description

This project presents a novel statistical and variational multiphase level set (SVMLS) approach to simultaneous bias correction and tissue segmentation for MRI image. The smoothness of the bias field is intrinsically ensured by the normalized convolution without any extra costs, which makes our method well fitted for images of various modalities, such as 3T and 7T MRI images. Moreover, the proposed SVMLS algorithm is robust to the initializations, therefore allowing for fully automatic applications. Comparisons with state-of-the-art method on real MRI images show the advantages of the proposed algorithm.

Structural patch decomposition based multi-exposure image fusion

Noise reduction for magnetic resonance images via adaptive multiscale products thresholding