一种基于贝叶斯和神经网络的医学图像组合分类方法

医学图像分类是当前医学图像自动诊断和模式识别领域的一个新的研究热点,其任务是从给定的医学图像训练样本中提取能反映图像内容的特征,并根据这些特征进行图像分类,实现医学图像中病变组织的自动识别,以保证临床医学诊断更客观、准确和科学.通过对医学图像分类中的一些关键问题分析和研究,提出一种基于贝叶斯和神经网络的医学图像组合分类方法,并据此构造出医学图像组合分类器.这种组合分类器能够充分发挥各个分类器的优点,获得较好的图像分类结果.

A Method of Medical Images Combining Classification Based on Bayesian and Neural Network

Medical images classification is a new research hotspot for medical image diagnosis automatically and pattern recognition. Its main tasks: firstly,to extract features,which can describe image contents,from training sample image sets; secondly,to classify the test image sets according to those features; finally,to recognize pathological tissue automatically by the classification results and can ensure a more objective and accurate result of clinic medical diagnosis scientifically. In this paper,some key prob...