论文部分内容阅读
目的检测甲状腺癌患者血清蛋白质,筛选特异的蛋白质标记物,构建用于甲状腺癌早期诊断的血清蛋白质指纹图谱模型。方法应用表面增强激光解吸电离主飞行时间质谱(SELDI-TOF-MS)技术测定81例血清标本(其中甲状腺癌40例,甲状腺腺瘤9例,健康人32例)的蛋白质质谱,用随机抽取的66例标本(甲状腺癌32例,甲状腺腺瘤9例,健康人25例)作为训练组,应用支持向量机进行训练和交叉验证,建立甲状腺癌诊断模型。结果区分甲状腺癌和正常人的诊断模型经留一法交叉检验该模型敏感性87·5%,特异性80%,用15例未知血清经盲法测试其敏感性为100%,特异性为86%;区分甲状腺癌和甲状腺腺瘤的诊断模型经留一法交叉检验敏感性为96·8%,特异性为89%。区分乳头状甲状腺癌和其他病理类型的甲状腺癌的诊断模型对乳头状甲状腺癌的判别率为97%,对其他病理类型的甲状腺癌的判别率为71%。结论表面增强激光解吸电离主飞行时间质谱技术结合支持向量机建立甲状腺癌血清蛋白质指纹图谱模型为早期筛查及诊断甲状腺癌提供了一种特异性强、敏感性高的新方法,值得进一步研究和应用。
Objective To detect serum protein in patients with thyroid cancer and screen specific protein markers to construct a serum protein fingerprint model for the early diagnosis of thyroid cancer. Methods The protein mass spectra of 81 serum samples (including 40 thyroid carcinomas, 9 thyroid adenomas and 32 healthy volunteers) were determined by surface enhanced laser desorption / ionization time of flight mass spectrometry (SELDI-TOF-MS) 66 cases (thyroid cancer 32 cases, thyroid adenoma 9 cases, 25 cases of healthy people) as a training group, the use of support vector machine for training and cross-validation, the establishment of thyroid cancer diagnosis model. Results The diagnostic model of thyroid cancer and normal people was divided into two groups. The sensitivity and specificity of this model were 87.5% and 80% respectively. The sensitivity and specificity of 15 cases of unknown serum were 100% and 86% respectively %; The diagnostic model that differentiated between thyroid cancer and thyroid adenoma was 96.8% with a specificity of 89%. Diagnostic models that differentiate papillary thyroid carcinoma from other pathological types of thyroid cancer were 97% for papillary thyroid carcinomas and 71% for thyroid carcinomas of other pathological types. Conclusions Surface enhanced laser desorption / ionization time-of-flight mass spectrometry combined with support vector machine to establish serum protein fingerprinting model of thyroid cancer provides a new method with high specificity and sensitivity for early screening and diagnosis of thyroid cancer. It deserves further study and application.