Font Size: a A A

Multimodal Imaging Study Of Tumor Of Nasal Cavity And Paranasal Sinuses

Posted on:2021-04-06Degree:MasterType:Thesis
Country:ChinaCandidate:W ZhangFull Text:PDF
GTID:2404330614463414Subject:Medical imaging and nuclear medicine
Abstract/Summary:PDF Full Text Request
Objective: To analyze the clinic value of CT and MRI multi-modal imaging diagnostic scheme on the detection of nasal and paranasal sinus tumors,so as to provide imaging basis for the preoperative evaluation and the determination of treatment scheme.Method: 70 nasal and paranasal sinus tumors surgically resected and pathologically confirmed in the Fourth Hospital of Zhangjiakou City from January 2016 to January 2018 were retrospectively analyzed as the research objects.CT,MRI plain scan,DWI and DCE-MRI were used for examination,and the imaging characteristics of CT,MRI plain scan,DWI and DCE-MRI multimodal were analyzed.The measurement indexes included 1)location,size,density,shape,boundary and metastasis in CT images.2)location,size,shape,boundary,surrounding node Structure relationship,lymph nodes and distant metastasis in MR images.3)Apparent Diffusion Coefficient(ADC)values in DWI measurement.4).Time Intensity Curve(TIC)typing in MRI DEC measurement,including mxslp,Tpeak and kep.According to geirnaerdt method,TIC curve can be divided into 3 types: type I(ascending type),type II(platform type)and type III(outflow type).Chi square test was used to compare the sensitivity and specificity of CT scan and MRI scan and curve type distribution,and t test was used to compare the ADC value of benign and malignant tumors.The parameters threshold,area under curve(AUC)and accuracy with statistical significance were determined by receiver operator characteristic curve(ROC).The sensitivity and specificity of different parameters to distinguish benign and malignant tumors were obtained.Results:There were 30 malignant tumors and 40 benign tumors.1.CT: the proportion of irregular shape in malignant tumors was 73.33%(22 cases),which was significantly higher than that in benign tumors(32.50%(13 cases),the difference was statistically significant(P < 0.05);the proportion of uneven density in malignant tumors was 80.00%(24 cases),which was significantly higher than that in benign tumors(7.50%(3 cases),the difference was statistically significant(P < 0.05);25 cases of bone destruction in malignant tumors were significantly higher than that in benign tumors.The difference was statistically significant(P < 0.05).2.MRI plain scan: the contrast-enhanced ratio of malignant tissue was 96.67%(29 cases),which was significantly higher than that of benign tissue(7.50%)(3 cases),the difference was statistically significant(P < 0.05).3.DWI: the average ADC value of malignant tumor was(0.901 ± 0.128)× 10-3mm2 / s,which was significantly lower than that of benign tumor(1.270 ± 0.197)× 10-3mm2 / s,the difference was statistically significant(P < 0.05).When the ADC value was 0.013mm2/s,the accuracy of diagnosis was 79.43%,the specificity was 68.95%,the sensitivity was 82.14%,and the AUC was 0.801.4.DCE-MRI: the proportion of TIC type in patients with malignant tumors is the highest,about 66.7%.The proportion of TIC type in patients with benign tumors is the highest,about 70.0%.The difference is statistically significant(P < 0.05).Compared with the parameters of mxslp,Tpeak and kep in DCE-MRI,the values of Tpeak in malignant tumor and benign tumor were(75.23 ± 19.55)s and(131.87 ± 42.15)s respectively,the difference was statistically significant(P < 0.05);the parameters of kep and mxslp in malignant tumor were(1.16 ± 0.37/min,18.73 ± 6.55)significantly higher than those in benign tumor(0.45 ± 0.24/min,11.02 ± 5.36),the difference was statistically significant(P < 0.0 5).5.The diagnostic specificity of DCE-MRI(77.50%)was significantly higher than that of CT(60.00%),the difference was statistically significant(P < 0.05);the diagnostic accuracy of DCE-MRI was 70.00%,the sensitivity was 60.00%,the diagnostic accuracy of CT was 62.86%,the sensitivity was 70.00%,the difference was not statistically significant(P > 0.05).When the diagnostic threshold of Tpeak was 84 s,the diagnostic accuracy was 71.43%,specificity was 77.50%,sensitivity was 63.33%,AUC was 0.764;when the diagnostic threshold of kep was 0.7/min,AUC was 0.897,diagnostic accuracy was 87.14%,specificity was 82.50%,sensitivity was 93.33%;when the diagnostic threshold of mxslp was 13.6,AUC was 0.851,diagnostic accuracy was 68.57%,specificity was 62.50%,sensitivity was 76.6 The AUC of ROC analysis was 0.932,the accuracy was 92.86%,the specificity was 90.00%,and the sensitivity was 96.67%.Among the four parameters,ADC,Tpeak and kep had better diagnostic efficacy,higher accuracy,specificity and sensitivity.The accuracy(92.86%),specificity(90.00%)and sensitivity(96.67%)of multivariate analysis of variance(Tpeak,ADC and kep)were significantly higher than those of univariate variance(P < 0.0 5).Conclusions:1.There are differences in TIC types between malignant tumors and benign tumors of nasal cavity and paranasal sinus.Most of malignant tumors belong to outflow type,while most of benign tumors belong to ascending type.2.The values of mxslp,Tpeak,kep and ADC can be used to differentiate and evaluate the benign and malignant masses of nasal cavity and paranasal sinuses,and the diagnostic accuracy of kep parameters is higher.The accuracy of multivariate parameters of MRI-DCE is obviously better than that of univariate variance.
Keywords/Search Tags:Nasal and paranasal sinus neoplasms, Computed tomography, Magnetic Resonance Imaging, Multiplanar reconstruction, Multimodal combined diagnosis
PDF Full Text Request
Related items