Multi-phase level set method for precise segmentation and correction of brain MRI
- PDF / 4,048,392 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 71 Downloads / 187 Views
ORIGINAL PAPER
Multi-phase level set method for precise segmentation and correction of brain MRI Yunyun Yang1
· Yunna Yang1 · Shenghua Zhong2
Received: 1 February 2019 / Revised: 9 March 2020 / Accepted: 4 June 2020 © Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract Medical image segmentation as an earlier application field in image segmentation is the key technology of medical image analysis and is also a key point and difficulty in clinical application. This paper proposes an accurate and robust active contour model based on the four-phase level set for medical MR images. First we define a new energy functional by combining the data term and the length term, where the data term is defined by transforming the energy functional of the multiplicative intrinsic component optimization (MICO) model into the level set framework after adding an edge detector function. Then, when we minimize the energy functional, we use the split Bregman method to improve the convergence speed. To test the performance of our model, we do lots of experiments according to the different brain MR images, which show that even under the severe influence of bias field or shadows, our model can still segment these images well, and our model is robust to the initial contours and noise. Moreover, our model is compared with the MICO model by experimental results and the numerical values, concluding that our model is better than the MICO model no matter in segmentation accuracy or in correction effect. Keywords Image segmentation · Bias correction · Split Bregman method · MR images Mathematics Subject Classification 90C47 · 65K10 · 49M37
1 Introduction Magnetic resonance imaging (MRI) is based on the principle that atomic nuclei with magnetic field spacing can produce transitions between energy levels under the action of magnetic field. MRI is helpful to examine the brain energy state and cerebral blood flow of epilepsy patients and has great diagnostic value for degenerative diseases. Compared with other imaging methods, the anatomical structure revealed by MRI is very realistic, and the lesion image is shown on the basis of a good and clear anatomical background, which makes the relationship between the lesion and the anatomical structure more clear. There are many articles
B
Yunyun Yang [email protected] Yunna Yang [email protected] Shenghua Zhong [email protected]
1
Harbin Institute of Technology, Shenzhen, China
2
Shenzhen University, Shenzhen, China
[1,2,9,12,16,20,24] to deal with magnetic resonance (MR) images, where the first five articles mainly talk about the image segmentation while others are mainly for the bias field correction. With the development of computer technology, a large number of traditional image processing methods and models have been widely used in medical image automatic segmentation, such as threshold methods, region growing methods, statistical models, the active contour models, clustering methods, and so on. [13,26] Chan and Vese [3] proposed the classical active contour model base
Data Loading...
