From the Intervention Centre (K.E.E., A.B.), Department of Radiology (P.D.T., J.K.H.), and Department of Neurosurgery (T.R.M.), Oslo University Hospital, N-0027 Sognsvannsveien 20, 0372 Oslo, Norway; Department of Radiology and Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Boston, Mass (K.E.E., M.C.P., O.R.); Department of Radiology, University of Texas Southwestern Medical Center, Dallas, Tex (M.C.P.); Department of Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Heidelberg, Germany (F.G.Z., L.R.S.); and Department of Physics, University of Oslo, Oslo, Norway (A.B.).
Radiology. 2015 Apr;275(1):228-34. doi: 10.1148/radiol.14140770. Epub 2014 Dec 8.
To develop a generic support vector machine (SVM) model by using magnetic resonance (MR) imaging-based blood volume distribution data for preoperative glioma survival associations and to prospectively evaluate the diagnostic effectiveness of this model in autonomous patient data.
Institutional and regional medical ethics committees approved the study, and all patients signed a consent form. Two hundred thirty-five preoperative adult patients from two institutions with a subsequent histologically confirmed diagnosis of glioma after surgery were included retrospectively. An SVM learning technique was applied to MR imaging-based whole-tumor relative cerebral blood volume (rCBV) histograms. SVM models with the highest diagnostic accuracy for 6-month and 1-, 2-, and 3-year survival associations were trained on 101 patients from the first institution. With Cox survival analysis, the diagnostic effectiveness of the SVM models was tested on independent data from 134 patients at the second institution.
were adjusted for known survival predictors, including patient age, tumor size, neurologic status, and postsurgery treatment, and were compared with survival associations from an expert reader.
Compared with total qualitative assessment by an expert reader, the whole-tumor rCBV-based SVM model was the strongest parameter associated with 6-month and 1-, 2-, and 3-year survival in the independent patient data (area under the receiver operating characteristic curve, 0.794-0.851; hazard ratio, 5.4-21.2).
Machine learning by means of SVM in combination with whole-tumor rCBV histogram analysis can be used to identify early patient survival in aggressive gliomas. The SVM model returned higher diagnostic accuracy values than an expert reader, and the model appears to be insensitive to patient, observer, and institutional variations.
利用基于磁共振成像的血容量分布数据开发一种通用支持向量机（SVM）模型，以预测术前胶质瘤患者的生存情况，并前瞻性评估该模型在自主患者数据中的诊断效能。
本研究经机构和地区医学伦理委员会批准，所有患者均签署了知情同意书。回顾性纳入 235 例来自 2 家机构的成年患者，这些患者术前均接受了 MRI 检查，术后经组织学证实为胶质瘤。应用 SVM 学习技术对基于 MRI 的全肿瘤相对脑血容量（rCBV）直方图进行分析。在第一家机构的 101 例患者数据上应用 SVM 算法，训练出预测 6 个月及 1、2、3 年生存率的最优诊断效能 SVM 模型。利用 Cox 生存分析，在第二家机构的 134 例独立患者数据上对 SVM 模型的诊断效能进行测试，这些数据与生存分析相关，且不受已知的生存预测因素（包括患者年龄、肿瘤大小、神经状态和术后治疗）的影响，并与专家阅片结果进行比较。
SVM 模型的预测结果与患者的生存情况显著相关（P＜0.001），且在调整了已知的生存预测因素后仍具有统计学意义。SVM 模型预测的生存情况与专家阅片结果具有较好的一致性（Kappa 值为 0.642-0.824）。
SVM 结合全肿瘤 rCBV 直方图分析的机器学习方法可用于识别侵袭性胶质瘤患者的早期生存情况。与专家阅片相比，SVM 模型具有更高的诊断效能，且对患者、观察者和机构差异不敏感。
