Dynamic Comparison of the Two Models to Enhance the Application Value of MRI in Cervical Cancer Case Classification

Objective:To explore the application of dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI)with Extended Tofts Linear(ETL)model and extended tofts(ET)model in pathological grading of cervical cancer.Method:Retrospective analysis was performed on 31 cases of patients admitted to Shaoxing People's Hospital from January 2016 to October 2019 who were pathologically confirmed to have cervical cancer.They underwent preoperative multistage pelvic DCE-MRI examination.Extended Tofts Linear and Extended Tofts hemodynamic model of cervical cancer were calculated separately,and the quantitative perfusion parameters including transport capacity constant(K^(trans)),the rate constant(K_(ep)),blood vessels,extracellular clearance volume fraction(V_(e)),blood vessels,and clearance volume fraction(V_(p)).According to the postoperative pathologic stage can be divided into low,medium and high differentiate cervical cancer group,the difference contrast of three groups of cervical cancer perfusion parameter were analyzed.The parameters with statistical significance were screened out,and the ROC curves for the differentiation of cervical cancer with different degrees were drawn.Result:The diagnostic efficacy of the quantitative parameters obtained by the two models in the pathological grading of cervical cancer was compared.The Ktrans values of the Exte Nded Tofts Linear model highly differentiated(0.518±0.180)min^(-1),poorly differentiated(1.032±0.408)min^(-1)and the Extended Tofts model highly differentiated(0.525±0.283)min^(-1),poorly differentiated(1.487±0.991)min^(-1)were significantly different between the highly differentiated and poorly differentiated groups(P<0.05).There were no significant differences in Kep,Ve and Vp between high and low differentiated cervical cancer groups(all P>0.05).There were no significant differences in Ktrans,Kep,Ve and Vp between the two models(all P>0.05).The AUC of Ktrans in the Exte Nded Tofts Linear model was greater than that of Ktrans in the Extended Tofts model for the identification of highly and poorly differentiated cervical cancers.Joint diagnosis:Logistic regression analysis was used to calculate the joint diagnostic probability values of Ktrans of the Extended Tofts Linear and Extended Tofts models,and the results showed that the AUC of the two models was greater than that of the single model.Conclusion:The Ktrans obtained by the Extended Tofts Linear model and the Extended Tofts DCE-MRI model had certain value for the pathological grading of high and low differentiated cervical cancer,and the overall efficacy of the former was better than the latter.Meanwhile,the differential value of the two models combined for the pathological grading of cervical cancer was better than that of the single model perfusion parameter,and the differential diagnosis sensitivity and specificity of the two models combined with Ktrans were the highest.