摘要
总结光学相干层析(OCT)医学图像的分割要求,对具有代表性的OCT医学图像分割方法进行详细论述,最后对OCT医学图像分割方法的发展趋势做出展望。代表性的OCT医学图像分割方法主要包括阈值分割方法、区域生长法、基于统计学的方法、基于活动轮廓的方法、基于图论的方法和形态学的方法等。综合利用多种医学图像信息,有效结合多种分割方法,注重提高方法的实时性、鲁棒性、精确性和自动化,将是OCT医学图像分割发展的重要趋势。
The requirements of optical coherence tomography（OCT） medical image segmentation are summarized in the paper, presenting a detailed review of the typical methods for OCT medical image segmentation and describing the prospect of the future research on the method for OCT medical image segmentation. The typical methods for OCT medical image segmentation mainly includes threshold segmentation method, region growing method, statistical method, deformable method, graph cut method, mathematical morphology and so on. Comprehensively utilizing the multiple medical image information, effectively combining various segmentation methods, and focusing on the improvement of the real- time performance, robustness, accuracy and automation of the method are the important development trends of OCT medical image segmentation.
出处
《中国医学物理学杂志》
CSCD
2016年第7期697-699,共3页
Chinese Journal of Medical Physics
基金
国家自然科学基金(21105127)
关键词
光学相干层析成像
医学图像
分割
医学图像特征
综述
optical coherence tomography
medical image
segmentation
medical image characteristic
review
