Application Of Apparent Diffusion Coefficient Histogram Map In The Evaluation Of Histopathological Features Of Endometrial Cancer

Objective:To explore the value of histogram analysis of apparent diffusion coefficient maps basing on entire tumor in assessing different histopathological features of endometrial cancer and identifying the benign and malignant tissues of endometrium.Methods:Patient group included 51 patients with endometrial cancer who were proved by post-operative pathology in the first hospital of Ji Lin University from May 2014 to December2016.Control group included 30 cervical cancer patients with a normal endometrial in the same period.All patients underwent conventional magnetic resonance imaging?MRI?,diffusion weighted imaging?DWI?and contrast enhanced magnetic resonance imaging?CE-MRI?before the operation.The obtained ADC maps were transmitted to the postprocessing software?Siemens Syngo?,region of interest?ROI?were drawn by freehand tools in each slices in which the tumor was visualized or the whole normal endometrial,and corresponding ADC histograms were automatically generate.Next,the ADC histograms were transformed into frequency distribution table and input into SPSS 20.0 software to reconstruct the ADC histogram of the whole tumor or normal endometrium and analyze the data.The patient group were divided according to the pathologic type,histologic grade,depth of myometrial infiltration,presence of cervical invasion,comparing the ADC values among groups and the patient group with control group,the ADC histogram parameters include mean,5th percentile,10 th percentile,25 th percentile,50 th percentile,75 th percentiles,90 th percentile and 95 th percentile.One-way analysis of variance was used to test differences in ADCs between four groups and independent t-testing was used to test differences between two groups.A value of P ? 0.05 was considered to be statistically significant.ROC curveanalysis was performed to determine the optimum threshold value for each parameter,and the corresponding sensitivity and specificity.Results:1.The results of histogram analysis of apparent diffusion coefficient maps in determining different histopathological characteristics of patient group:?1?There was a significant difference in 5th percentile,10 th percentile and 25 th percentile of ADC values between Grade1,Grade 2,Grade 3,and non-endometrioid tumors?P<0.05?;?2?The 5th percentile ADC values of endometrioid tumors?G1+G2+G3?showed significantly lower than non-endometrioid tumors?P<0.05?,the area under the curve was 0.654,and the sensitivity,specificity was0.29,1.00 respectively when set the cut-off value to 719.00x10^-6mm²/s;?3?The mean,5th percentile,10 th percentile,25 th percentile,50 th percentile of ADC values showed significantly lower for tumors with deep myometrial invasion than superficial or no myometrial invasion?P<0.05?.The optimum cut-off value of 5th percentile showed the highest area under the curve?AUC=0.725?for determining deep myometrial invasion,corresponding sensitivity and specificity is 0.51,1.00 respectively;?4?All parameters observed showed no significantly correlation for where there is cervical involvement or not?P>0.05?.2.The mean,5th percentile,10 th percentile,25 th percentile,50 th percentile,75 th percentiles,90 th percentile and 95 th percentile of ADC values showed significantly differences between patient group and control group?P<0.05?.The 90 th percentile have the highest diagnostic efficiency with a highest area under the curve?AUC=0.987?,applying a cut-off value of less than 1438.90x10^-6mm²/s,we obtained a sensitivity,specificity for malignancy of0.97 and 0.94.Conclusions:1.Histogram analysis of ADC maps based on entire tumor can be useful for predicting endometrial cancer patients with high risk factors,such as lower grade,non-endometrioid,and deep myometrial invasion,but not for cervical involvement.2.Histogram analysis of ADC maps can be useful to differentiating malignant from benign lesion of endometrium,the 90 th percentile of ADC was considered the most promising parameter.3.Histogram analysis of ADC maps can be used for quantitative analysis of lesions to provide more accurate information for clinicians.