核磁共振成像(MRI)以其无辐射、多方位成像、空间分辨率高等优点在影像医学领域广泛应用,核磁共振图像的分割也发挥着越来越重要的作用。对应用较广的核磁图像的分割算法的原理和应用进行了系统的综述,将核磁图像分割算法分为5个主要研究方面:基于阈值,基于模式识别,基于活动轮廓模型,基于马尔科夫随机场(MRF),基于图切割;给出了不同算法分割特点和相关应用范围,并将部分算法应用在腹部核磁图像上进行分割实验,展示了不同算法分割核磁图像后的效果和特点。最后,展望了核磁图像分割的未来的发展趋势。
Magnetic resonance imaging (MRI) plays a more and more important role in medical image area for its advantages of nonradiative, multiple imaging and high spatial resolution. This review gives a systematic discussion over a couple of MRI segmentation algorithms that are used widely to help people have an entire knowledge of MRI segmentation methods. On the base of classification and summary of MRI segmentation, the MRI segmentation algorithms have been classified into 5 different categories after preliminary investigation and survey. Based on threshold, pattern recognition, active contours, Markov random field (MRF) and graph cut, separately. After further investigation and survey we summarize the characters and field of applications of the 5 different algorithms, then take a few segment experiments among these algorithms on abdomen MRI and present their distinct characteristics. At last, we take a prospect at the future of the MRI segmentation.