The Application Of MR High-angular-resolution Diffusion Imaging In Glioma Peritumoral Edema

Backgrond and purpose:Cerebral glioma(CG)is the most common malignant tumor in the skull,with the characteristics of high invasiveness,high lethality and easy recurrence.The glioma grows infiltratively,and most of them confined to the surrounding brain tissue,and the boundary with the surrounding normal brain tissue is unclear.Peritumor edema(PTE)is one of the main biological behavior characteristics of gliomas,which has a significant impact on clinical prognosis,and it is also one of the important reasons for the high morbidity and mortality of gliomas.A major goal of modern surgical treatment of brain tumors is to maximize tumor removal while minimizing damage to critical areas of the brain.Magnetic Resonance Imaging(MRI)has a high tissue resolution,which not only can non-invasively assess tissue microstructure,cell and blood vessel proliferation,and capillary permeability,but also can visually show the deep gray matter and white matter fibers of tumor Conducive to surgical treatment.Diffusion tensor imaging(DTI)is an imaging method that uses changes in magnetic resonance signals caused by the anisotropy of water molecules in tissues to detect tissue microstructures.It is mainly used to study the anatomical nerve bundles of the central nervous system and display of anatomy of white matter fibers.High-angular-resolution diffusion imaging(HARDI)technology is based on DTI.Compared with the fiber bundle tracking technology based on DTI data,HARDI can solve the problem of fiber bundle crossing more accurately.Compared with the previous algorithm,the tracking result has a qualitative leap in accuracy and precision.For HARDI imaging,Q-ball imaging(QBI)technology,high-order tensor model(High Order Tensor,HOT),spherical harmonic(SH)and spherical deconvolution(SD)and other technical methods are mostly used.SD can be used to calculate the fiber orientation distribution(FOD)in each voxel.FOD contains different fiber directions,including the uncertain directions surrounding them,and their respective volume fractions.A definite study found that the HARDI model has advantages in the evaluation of high-grade gliomas with severe edema,and is particularly suitable for situations where complex fiber bundles need to be reconstructed.From the literature in recent years,HARDI fiber bundle tracking will become a relatively active field of research and development.The purpose of this study is to explore the application of HARDI model parameter value of quantitative analysis in the periglioma edema area;and the quantitative analysis of the correlation between the degree of glioma peritumoral edema and gender,age,glioma grade,IDH classification.Materials and Methods:The imaging and pathological data of 147 patients with glioma confirmed by surgery and pathology were collected,and the patients were scanned using SIEMENS Prisima 3.0 T magnetic resonance scanning.All patients underwent conventional MRI plain scan,HARDI and dynamic enhancement scan.All plain scans and enhanced images of 147 patients with pathologically confirmed glioma were imported into RadiAnt DICOM Viewer software for observation and delineation of edema area.Bitzer's method was used to calculate edema index(EI),Gender,age and the correlation of glioma grade and IDH classification.The images of patients with EI>3.0 and pathologically proven IDH wild-type glioblastoma with obvious PTE(55 cases)were selected and transmitted to the syngo.via Frontier framework workstation.Processing,then get Track-density imaging(TDI)map,Track-density tractograpy(TDT)map.Locating in the MR Basic window according to the T2WI positioning,manually draw ROI on the TDI map to measure the track density of the near area of tumor PTE,the far area of tumor PTE,suspicious white matter area,normal white matter area on the tumor side and the corresponding normal white matter area on the opposite side of the same level(The track density value on the tumor side)/(the normal white matter area on the contralateral side)gives the corrected track density.SPSS 20.0 software was used for statistical analysis of each parameter.Use histogram to analyze the distribution of gender and age in various types of gliomas.The EI value was compared between the groups using the Kruskal-Wallis rank sum test or T test.P<0.05 was considered statistically significant.For further pairwise comparison,the L-S-D test was used,and P<0.05 was considered statistically significant.One-way ANOVA was used to determine the difference in the corrected track density in the PTE area of the glioma,and the L-S-D test was used to compare the two groups.Draw the ROC curve of effective parameters,and calculate the area under the curve,then predict the corresponding threshold?sensitivity?specificity and the Youden index.Results:In this study,the age of glioma patients ranged from 18 to 70 years,with an average age of 46.80±12.44 years,male:female=11:10,and there was no significant difference in gender in the overall distribution.There were statistically significant differences in age between high/low grade groups,WHO ?/?groups,WHO ?/? groups and IDH mutant/IDH wild type groups(P=0.004,P=0.000,P=0.000,P=0.000).There was no significant difference in EI value between age and gender groups(P=0.290,P=0.453);The difference between high/low grade groups,WHO ?/? grade groups,WHO ?/? grades and IDH wild type/mutant groups was statistically significant(P=0.000,P=0.000,P=0.041,P=0.005).The area under the curve(AUC)of the EI value in the differential diagnosis of the high-level and low-level groups was 0.767,which was the largest,and the threshold was 1.630,and the sensitivity and specificity were 77.0%and 62.7%,respectively.The difference in the corrected track density between the near area tumor PTE,the distant area of tumor PTE,relative normal white matter area,and normal white matter area was statistically significant(F=229.061,P=0.000).In further pairwise comparison,The differences of each group were statistically significant(all P<0.001).The AUC of the corrected track density in the differential diagnosis of the near area of tumor PTE and the relatively normal white matter area was 0.992,the threshold was 0.578,and the specificity and the sensitivity were 92.7%and96.4%.Conclusion:1.The EI value is different in the glioma grade and IDH classification.2.The HARDI parameter track density is characterized by a gradual change in various areas around the glioblastoma,which reflects the tumor invasion of the glioblastomat.