X线纹理分析鉴别诊断乳腺叶状肿瘤与纤维腺瘤

X-ray texture analysis in differential diagnosis of phyllodes tumor and fibroadenoma in breast

目的探讨X线纹理分析鉴别诊断乳腺叶状瘤(BPTs)与乳腺纤维腺瘤(BF)的价值。方法回顾性分析32例BPTs和32例BF患者的术前乳腺X线片。采用MaZda 4.6软件手动勾画病灶最大层面ROI。分别采用Fisher系数、聚类偏差概率结合平均关联系数(POE+ACC)、协同信息(MI)及联合应用上述3种方法(MI+PA+F)选择ROI内最具鉴别价值的纹理特征参数,并与术后病理结果对照,比较影像医师主观阅片与采用纹理分析诊断BPTs及BF错判率的差异。结果影像医师鉴别BPTs与BF的错判率为20.31%(13/64);采用Fisher系数进行纹理特征分析的错判率为14.06%(9/64),POE+ACC为23.44%(15/64),MI为25.00%(16/64),MI+PA+F为3.12%(2/64)。MI+PA+F的错判率低于影像医师主观阅片及POE-ACC、MI(P均<0.05)。结论 X线纹理分析可用于鉴别BPTs和BF。

Objective To investigate the value of X-ray texture analysis in differential diagnosis of breast phyllodes tumors(BPTs)and breast fibroadenoma(BF).Methods Preoperative X-ray data of 32 patients with BPTs and 32 patients with BF were retrospectively analyzed.MaZda 4.6 software was used to manually draw ROIs on the maximum level of mammography images.The optimized texture parameters of texture analysis were selected with Fisher coefficient,probability of classification error and average correction coefficient(POE+ACC),mutual information(MI)as well as combination of the above three methods(MI+PA+F),respectively.The diagnostic results obtained by radiologist or using texture analysis techniques were contrasted with pathology.The misclassification rates of BPTs and BF were compared between texture analysis and subjective diagnosis by radiologist.Results The misclassification rate of subjective diagnosis,Fishers,POE-ACC,MI and MI+PA+F was 20.31%(13/64),14.06%(9/64),23.44%(15/64),25.00%(16/64)and 3.12%(2/64),respectively.The misclassification rate of MI+PA+F was lower than that of subjective diagnosis,POE-ACC and MI,respectively(all P<0.05).Conclusion It is feasible to use texture analysis on X-ray films for differentiation of BPTs and BF.