【作者】 马琳；

【导师】 宋志坚；

【作者基本信息】 复旦大学 ， 医学信息学， 2012， 博士

【摘要】 脑胶质瘤是中枢神经系统最常见的原发恶性肿瘤,在颅内肿瘤的发病率中可算首屈一指。近年来,其临床诊断、治疗策略和基础研究方面都有了很大的进展,从最初手术禁区到实现单纯外科切除；从保护重要脑功能区前提下的手术切除,到如今的手术、放疗和化疗等综合治疗策略。随着相关领域如计算机科学、数字图像标准(DICOM)和医学影像学的迅速崛起,尤其是数字医学和医学信息学的兴起,将计算科学和信息科学成功地运用于医学信息的获取、表达、处理、存储、传输以至于决策支持系统的研究,于是带动了磁共振技术从单纯人体器官解剖结构的显示过渡到功能影像的表达,从计算机辅助导航定位系统到基于影像的导航神经外科的出现,人们对胶质瘤的认识也逐渐深入,无论是术前影像诊断、外科手术入路和质量、放化疗技术的精准性和基础研究等方面都有了极大的改观。尽管如此,相对于全身其他系统肿瘤来说,胶质瘤病人的死亡率和复发率仍然是居高不下,而且高死亡率、高复发率都和胶质瘤级别密不可分。胶质瘤研究领域,为近年来诸多学者们所关注,随着弥散加权成像(DWI),弥散张量成像(DTI),磁敏感加权成像(SWI)、磁共振波谱(MRS)等新技术的开展,数字图像处理和影像学向着微观世界和功能显像倾斜,特征影像和微观病理相关性研究,大脑白质纤维失踪技术(FT),血氧水平依赖的脑功能区显示技术(BOLD)、基于影像的神经外科导航技术(IGNS)等等已经初步应用于临床,但是对于利用神经导航技术和弥散张量成像进行胶质瘤的术前分级诊断,却少有报道。高低级别胶质瘤在手术切除范围和综合治理策略方面都大不相同,可以说预后与手术效果直接相关,目前术前诊断依旧是依靠临床医生和神经影像专家的共同判断,最终确诊还得等到术后病理组织切片,对其术前无创诊断和定量分析方面尚未见报道,本文对此开展了相关的研究。目的：回顾性地分析弥散张量成像(Diffusion Tensor Imaging, DTI)的两个常用参数,部分各向异性(fractional anisotropy, FA)、表观弥散系数(Apparent Diffusion Coefficient, ADC),以及DTI形态参数,线性各向异性系数(Linear Anisotropy Coefficients, CL)、平面各向异性系数(Planar Anisotropy Coefficients,CP)、各向同性系数(Spherical Anisotropy Coefficients, CS)等参数值能否用来表现高低级别胶质瘤本身的不同区域的结构和功能状态；定量分析每个独立的DTI参数是否能够和高低级别分组之间具有相关性；研究DTI参数是否能够用来建立高低级别胶质瘤的鉴别诊断模型。资料与方法：对15例高级别和10例低级别胶质瘤病人的常规MRI图像T1WI、T2WI (b=0)和DTI张量图像进行配准融合,在相应部位放置感兴趣区,分别测量肿瘤实质部分和瘤周水肿浸润区的DTI参数值FA、ADC、CP、CS、CL、方法一：使用两个相关样本的Wilcoxon检验,分别分析高级别和低级别胶质瘤,在肿瘤的不同区域(肿瘤实体和瘤周浸润区)的5个DTI参数值的差异,统计学显著性检验水平为P<0.05；方法二：使用Mann-Whitney U检验法,比较5个DTI参数作为独立因素在高低级别分组中的差异有无统计学意义(P<0.05)；方法三：先用单因素Logistic回归分析,进行变量筛选；再用多因素的Logistic回归分析用来拟合高低级别胶质瘤的最佳分类模型。用受试者工作特征曲线(Receiver Operating Characteristic Curve, ROC)来评估不同DTI参数在诊断高低级别胶质瘤方面的效能,取Youden指数最大时为分级的临界值。结果：1.无论是高级别,还是低级别胶质瘤,FA、CL、CP、CS值在肿瘤实质部分和瘤周浸润区的差异均有显著的统计学意义,其低级别组的P值分别为0.004、0.014、0.017、0.003,高级别组的P值分别为0.001、0.001、0.005、0.002；但是ADC值在两组肿瘤的不同区域的差异并无统计学意义(P=0.457和0.376)。其中,在两组肿瘤中表现相同的是,肿瘤实质部分与瘤周浸润相比较,FA、CL、CP值增加,而CS值则降低。但是ADC值在不同级别的肿瘤中却有着相反的表现,在低级别胶质瘤中,肿瘤实质区ADC值比瘤周浸润区的有所升高；而高级别的胶质瘤中,实质区ADC值则要略低于瘤周浸润区,尽管这种差异在P<0.05的检验水平上没有显示出统计学意义。2. FA、CP、CL值由瘤体区向外呈递增趋势,在低级别胶质瘤中,FA增幅最大,为78.40%,FA的最大值0.45(瘤周浸润区),最小值0.07(肿瘤实质区)；在高级别胶质瘤中,CL的增幅最大,为117.91%,CL的最大值0.40(瘤周浸润区),最小值0.03(肿瘤实质区)。而对于CS值而言,在低级别和高级别胶质瘤中由瘤体向外呈递减趋势,瘤周水肿区CS值均显著低于瘤周浸润区,分别降低了17.76%和21.11%。3.分析胶质瘤的不同区域每个DTI参数结果显示,虽然高低级别分组之间有不同程度的升高或者降低,但是这种差异在本次研究中,P<0.05的检验水平上没有统计学意义；4.多因素的Logistic回归分析表明,结合的多参数Logistic回归模型能够有效区分高、低级别胶质瘤,包含CS、FA和ADC三个参数的拟合方程为最佳预测模型,敏感度86%,特异度80%,ROC曲线下面积0.82。结论：1.无论是高级别还是低级别胶质瘤,DTI参数在它们的实质部分和瘤周浸润区的表现有显著差异,在很大程度上反映了脑组织和肿瘤组织在结构和功能上的不同,可用于量化分析肿瘤不同部位的解剖结构；这一点也是和我们传统的微观病理相符合的；至于是否和脑肿瘤的浸润程度相关,需要进一步的研究,不同部位的FA、CL、CP、CS等参数值可帮助显示肿瘤的不同区域,在神经导航过程中对于指导脑胶质瘤的手术切除范围有一定的意义。2.单个的DTI参数,可能无法进行高级别和低级别胶质瘤之间的准确区分；关于这一点,有待于更大样本量的后续验证,得出更为稳定的结果。3.DTI多参数模型是术前进行高低级别胶质瘤鉴别诊断的一个有效方法。更多还原

【Abstract】 Glioma, the most common primary malignant tumors of the central nervous system, can be considered second to none in the incidence of intracranial tumors. In recent years, the clinical diagnosis of gliomas, treatment strategies and basic research have a great deal of development. The intracranial tumor is from considering as just the restricted area of the surgery in the early to taking simple surgical excision; Treatment strategies are also from total removal on the basis of protecting critical brain function areas to nowdays multimodal treatment including surgical resection, radiotherapy, and chemotherapy. With the rapid rise of computer science, digital image standards and medical imaging, especially magnetic resonance imaging from simple anatomical structure to functional imaging, computer-aided navigation and positioning system, image-guidance neurosurgery[1] and other emerging areas of research on glial gradually in-depth understanding of tumor, preoperative imaging diagnosis, surgical approach successfully, the precision of radiotherapy and chemotherapy techniques and basic research have been greatly improved.Although the multimodal treatment of gliomas in recent years has made great progress, relative to the low-grade gliomas, the prognosis of patients with high-grade gliomas remains poor[2-31and most of their survival time is less than1year. Treatment strategies including surgical resection, radiotherapy, and chemotherapy for these two grades are very different. Therefore, it is crucial to discriminate accurately between high-grade and low-grade gliomas preoperatively.To date, glioma grading is still limited to subjective observation of both neurosurgeons and neuroradiologist without accurate quantitative analysis methods. Therefore, we studied diffusion tensor characteristics of high-grade and low-grade gliomas based on various areas of the tumor.OBJECTIVE:To ascertain whether diffusion tensor imaging (DTI) metrics including tensor shape measures such as planar and spherical isotropy coefficients (CP and CS) can be used to distinguish high-grade from low-grade gliomas. METHODS:Twenty-five patients with histologically proved World Health Organization (WHO) grade I-IV gliomas (10low-grade and15high-grade) were included in this study. Contrast-enhanced Tl-weighted images, non-diffusion weighted b=0(b0) images, fractional anisotropy (FA), apparent diffusion coefficient (ADC), CS and CP maps were co-registered and each lesion was divided into two regions of interest (ROI):enhancing and immediate peritumoral edema (edema adjacent to tumor). Univariate and multivariate logistic regression analyses were applied to determine the best classification model.RESULTS:There was a statistically significant difference in the multivariate logistic regression analysis. The best logistic regression model for classification combined three parameters (CS, FA and CP) from the peritumoral part (p=0.024), resulting in86%sensitivity,80%specificity and area under the curve of0.81.CONCLUSION:Our observations revealed that combined DTI metrics can function in effect as a non-invasive measure to distinguish between low-grade and high-grade gliomas.更多还原