论文部分内容阅读
一个超声波图像评价系统,此系统的目的在于能在临床上辅助诊断慢性肝疾病,此系统使用肝疾病严重程度的数字化得分来反映慢性肝疾病的进展程度。慢性肝脏疾病的进展主要与软组织的纤维化数量有关,一种有效的评测指标称为计算机形态测量得分(CM),CM能正确地测量人体肝脏切片中肝组织的纤维化比例,通常被认为是评价慢性肝疾病的“金标准”。在肝疾病超声图像的纹理特征与相应地CM得分之间,使用径向基函数网络(RBF)来建立两者之间的相关性。RBF网络的输出即可指示出肝病严重度超声波图像得分(UDS)。在120个测试图像中使用UDS评分标准,正确分类比为可达92.5%。此结果显示出,UDS标准极有可能为诊断慢性肝疾病提供一个重要的参考标准。
An ultrasound image evaluation system designed to assist clinically in the diagnosis of chronic liver disease. The system uses digitalized scores of severity of liver disease to reflect the progression of chronic liver disease. The progress of chronic liver disease is mainly related to the amount of soft tissue fibrosis. An effective measure called the computer morphology measurement score (CM), CM can correctly measure the hepatic fibrosis in human liver sections, which is usually considered as Evaluation of the “golden standard” for chronic liver disease. Between the texture features of the liver disease ultrasound image and the corresponding CM score, a radial basis function network (RBF) is used to establish the correlation between the two. The output of the RBF network is indicative of a liver disease severity ultrasound image score (UDS). Using the UDS score in 120 test images, the correct classification ratio was 92.5%. This result shows that the UDS standard is most likely to provide an important reference for the diagnosis of chronic liver disease.