论文部分内容阅读
目的 :研究大肠上皮异型增生多指标多参数检测及其在大肠肿瘤的计算机自动化诊断中的应用价值。方法 :本课题应用计算机图像分析技术对 5 3例大肠癌、30例大肠腺瘤及 10例正常肠粘膜进行DNA含量、核形态参数、AgNOR的计数、颗粒形态参数及PCNA等多指标多参数检测。结果 :多数参数在各组间具有显著性差异 (P<0 0 1) ,经判别筛选 ,DNA指数、AgNOR的平均颗粒数、阳性率、平均光密度、颗粒表面因子 ,PCNA的平均光密度、阳性率 ,核的形状因子、核宽径及平均光密度等 10个参数判别能力较大 ,为最佳判别参数。参数值大多随着正常、腺瘤及腺癌顺序递增 ,显示这些参数均能较好反映肿瘤的演进过程。PCNA大部分参数值在肿瘤的演讲过程中其高峰出现在腺瘤阶段 ,故其在大肠癌早期诊断中的价值需进一步探讨。结论 :应用图像分析技术对大肠癌及癌前病变进行多指标多参数的计量分析 ,具有鉴别诊断重要意义 ,为大肠肿瘤的自动化诊断提供客观依据。
Objective: To study the multi-parameter and multi-parameter detection of large intestine epithelial dysplasia and its application value in computer automated diagnosis of colorectal neoplasms. METHODS: In this study, computer image analysis techniques were used to detect DNA content, nuclear morphology, AgNOR counts, particle morphology parameters, and PCNA detection in 53 cases of colorectal cancer, 30 cases of colorectal adenoma, and 10 cases of normal intestinal mucosa. . Results: Most of the parameters were significantly different among the groups (P<0 01), and were identified and screened, DNA index, AgNOR average particle number, positive rate, average optical density, particle surface factor, PCNA average optical density, The positive rate, nuclear shape factor, nuclear width and average optical density and other 10 parameters have a greater ability to discriminate. It is the best discriminant parameter. Most of the parameter values increase with the order of normal, adenoma and adenocarcinoma, indicating that these parameters can better reflect the evolution of the tumor. Most of the parameters of PCNA appear in the adenoma stage during the presentation of tumors, so its value in the early diagnosis of colorectal cancer needs further exploration. Conclusion : The application of image analysis technology to the multi-indicator and multi-parameter analysis of colorectal cancer and precancerous lesions has important significance in differential diagnosis and provides an objective basis for the automated diagnosis of colorectal cancer.