文章摘要

文章信息/Info

Title:: Research of Brain Tumor Pathology Classification based on DTI Quantitative Parameters

Author(s):: LUO Ting1; TAO Ling1; YU Zhou1; WANG Kunsheng1; YAO Liuye1; College of Automation, Nanjing University of Aeronautics & Astronautics, Nanjing 211106, China

Keywords:: Magnetic resonance diffusion tensor imaging (DTI); Fractional anisotropy(FA); Apparent diffusion coefficient(ADC); Brain tumor classification; Multimodel image

摘要:: 探讨DTI在脑肿瘤分级中的应用价值，通过对FA值和ADC值的测量分析，发现DTI定量参数与脑肿瘤级别的关系，为临床应用提供信息。对19例经病理证实的不同等级脑肿瘤患者按肿瘤等级分组，对肿瘤实质区与对侧区正常脑组织的FA值和ADC值行配对〖WTBX〗t〖WTBZ〗检验，比较肿瘤实质区与对侧区正常脑组织的差异；另就高、低级别脑肿瘤FA值和ADC值做独立样本〖WTBX〗t〖WTBZ〗检验，分析差异。脑肿瘤患者组内肿瘤实质区较对侧正常脑组织FA值及ADC值均具有显著差异。低级别脑肿瘤较高级别脑肿瘤ADC值更低，且差异显著。脑肿瘤实质区FA值和ADC值可以帮助脑肿瘤区域与正常脑组织的界定，但对于脑肿瘤分级的应用价值尚有待考究。多模态图像的联合分析方法将成为脑肿瘤级别无创划分的新热点。

Abstract:: To investigate the value of DTI in brain tumor classification, the relationship between the DTI quantitative parameters and the level of brain tumor was studied by analysing the parameter values of FA and ADC, which can provide information for clinical application. 19 cases confirmed by pathology of tumor patients of different levels were divided into different groups according to the grades of tumor, the differences of the tumor nature area and the contralateral normal area were compared by using matching 〖WTBX〗t〖WTBZ〗 test about FA and ADC values, the differences of the high and low level of brain tumors were analyzed by using independent sample 〖WTBX〗t 〖WTBZ〗test about FA and ADC values. There were significant differences in FA and ADC values between the tumor nature area and the contralateral normal brain tissue of brain tumor patients. The ADC values of low levels of brain tumors were lower than that of high levels, and the differences were significant. Using FA and ADC values of brain tumor real area is helpful for distinguishing brain tumor areas and normal brain tissue, but whether it is effective for brain tumor classification remains to be confirmed. The conjoint analysis method of multimodal images will become a new hot spot of noninvasive brain tumor classification.