中心型肺癌的CT图象特征及病理分型的定量研究

为探讨提高CT影像对中心型肺癌病理类型的诊断准确率的方法,对1997年1月~2001年12月146例中心型肺癌患者的13个临床及CT图象特征逐一进行量化,利用计算机影像计量诊断技术,对影响因素进行单、多因素分析,并对另外14例患者进行前瞻性应用研究。建立了中心型肺癌的CT图象特征与病理分型的关系的数学模型,经验证其诊断符合率为57·1%。初步研究结果提示,计算机影像计量诊断技术对中心型肺癌的组织学分型有一定诊断价值。

Quantitative study of the relationship between CT image characteristics and histological pattern in central lung cancer

The objective of this study was to probe into the means for improving the diagnostic accuracy of pathological patterns in central lung cancer. Thirteen clinical and CT image characteristics in 146 patients with central lung cancer were quantified according to computer demand and were analyzed with single- and multi-factor correlation analysis. Another 14 patients were studied prospectively with the results. The mathematical model of the relationship between CT image characteristics and pathological pattern was set up and its diagnose accordance rate reached to as high as 57.1%. In conclusion, the computer quantitative technology has high diagnostic accuracy of pathological patterns in central lung cancer.