摘要
本研究应用图像分析技术测量了４８例（良、恶性各２４例）乳腺肿块细针穿刺涂片的细胞学参数，发现恶性组平均细胞核面积、周长、等效直径、圆度系数、各灰度参数、质量以及各参数标准差，均明显大于良性组；并应用ＳＡＳ软件包建立了Ｂａｙｅ＇ｓ判别模型，通过回代检验，灵敏度为９３．３３％，特异度为１００％。
The image analytic technique was utilized for measuring cyto logical parameters of 48 fine needle aspiration smears(benign 24,malignant 24).It was found that the mean nuclear area(MNA),mean nuclear perimeter (MNP),mean nuclear diameter(MND),coefficient of roundness,degree of gray,volume and the standard deviates of these parametets in malignant group were significantly higher than those in benign group.These parameters were dealed with SAS on computer,and the diagnostic system was built in Baye's differentiating model,and its sensitive rate is about 93.33% and specific rate is about 100%.
