支持向量机预测食管鳞癌患者术后生存期
DOI: 10.3971/j.issn.1000-8578.2015.08.004
刘瑞敏,
张天,
晁玮霞,
朱晗,
王广超,
王明,
谷娟,
贺红梅,
赵云岗,
韩大正,
马瑾,
马远方,
魏书堂,
齐义军
Keywords: 食管鳞状细胞癌,支持向量机,预后,14-3-3&sigma,热休克蛋白gp96,巨噬细胞移动抑制因子,Clinical Analysis of Radiotherapy Combined with Cisplatin-based Chemotherapy for Locally Advanced Esophageal Squamous Cell Carcinoma,Relationship Between NF-κB and Epithelial-to-mesenchymal Transition of Radiationinduced Human Esophageal Carcinoma Cells,Effect of Gene Silencing of RNF2 on Radiosensitivity of Esophageal Carcinoma Cells,Expression and Clinical Significance of Stem Cells Gene Lgr5 in Esophageal Squamous Cell Carcinoma Tissues,Department of Thoracic Surgery, Zhejiang Cancer Hospital, Hangzhou 310022, China,Effects on Cell Proliferation, Invasion and Migration of Esophageal Adenocarcinoma by Inhibition of Notch and PI3K /Akt Pathway,Epidemiologic Investigation on Mortality Due to Esophagus Cancer in Residents in Xiamen, 2005-2014,Effect of GANT61 on Apoptosis of Esophageal Squamous Carcinoma Cells and Its Mechanism,Prognosis Factors for Patients with Stage pT3N0M0 Thoracic Esophageal Squamous Cell Carcinoma after Radical Resection,Expression of EZH2 and H3K27me3 in Esophageal Squamous Cell Carcinoma and Their Significance,Expression of DMRT2 in Esophageal Squamous Cell Carcinoma Tissues and Related Clinical Significance,Relationship of Secreted Protein Acidic and Rich in Cysteine Protein Expression with Prognosis of Postoperative Patients with Esophageal Squamous Cell Carcinoma,Metastasis?associated Lung Adenocarcinoma Transcript1(MALAT1) Expression in Middle Thoracic Esophageal Squamous Cell Cancer Tissues and Its Relationship with Prognosis,Relationship Between Lymphocyte Nadir and Radiotherapy Efficacy or Prognosis of Patients with Esophageal Carcinoma,Comparison of 6th and 7th Editions of AJCC/UICC TNM Staging for Esophageal Cancer#br# Patients Received Radiotherapy
Abstract:
摘要 目的 应用支持向量机（support vector machine, SVM）建立食管鳞状细胞癌（esophageal squamous cell carcinoma, ESCC）术后生存期预测模型并评估该模型判断ESCC生存期的效能。方法 随访168例接受根治性手术治疗的ESCC患者，分析ESCC临床病理特征和14-3-3σ、热休克蛋白gp96、巨噬细胞移动抑制因子（migrationinhibitory factor, MIF）等3个蛋白的表达规律与ESCC生存期的相关性；应用Matlab软件进行SVM运算，对训练组128例ESCC患者建立最优预后分类模型ESCC-SVM，并用测试组40例患者验证分类效率，ROC曲线分析ESCC-SVM及其他预后相关因子对高低死亡风险ESCC的识别能力。结果 ESCC-SVM由性别、T分期、组织学分级、淋巴结转移、TNM分期、14-3-3σ和gp96等7个最优属性组成，该模型区分训练组和测试组ESCC五年整体生存率的最大AUC分别为0.96、0.86、准确率分别为97.7%、90.0%，明显优于目前临床应用的TNM分期（准确率分别为62.5%、67.5%）及其他各临床病理属性。Cox多因素比例风险回归模型分析发现年龄、T分期、gp96和ESCC-SVM是影响ESCC术后生存期的独立因素。ESCC-SVM与性别、T分期、组织学分级、淋巴结转移、TNM分期和14-3-3σ均显著相关。结论 本研究建立的ESCC-SVM为预后评估、临床治疗方案选择及个体化治疗提供了理论依据
Full-Text
