The Value Of Multi-parameter Magnetic Resonance Imaging Combined With Radiomics Of Primary Lesion Of Rectal Adenocarcinoma To Predict Non-enlarged Metastatic Lymph Node And Recurrence Before Operation

Objective:To investigate the value of intravoxel incoherent motion diffusion weighted imaging（IVIM-DWI）parameters combined with T₂WI texture analysis of primary lesions of rectal adenocarcinoma in preoperative prediction of lymph node metastasis with short diameter≤9 mm.Methods:Retrospective analysis was performed on 115 cases of rectal adenocarcinoma confirmed by surgical pathology in our hospital from June 2015 to October 2020.All patients received conventional rectal MRI and IVIM-DWI scan before surgery and underwent radical resection.According to the pathological results of lymph node,the patients were divided into lymph node metastatic（LN+）group（n=44）and non-metastatic（LN-）group（n=71）.IVIM-DWI parameters of primary rectal adenocarcinoma were measured including apparent diffusion coefficient（ADC）,diffusion coefficient（D）,pseudo diffusion coefficient（D^*）,and perfusion fraction（f）.The region of interest（ROI）of the whole lesion of rectal adenocarcinoma was delineated on axial T₂WI,and then the ROIs were imported into GE Analysis Kit software to extract 3D texture feature.The differences of IVIM-DWI parameters and texture feature parameters were compared by independent sample t test or Mann-Whitney U test,and the optimal texture feature parameters with independent predictive function were screened by multivariate logistic regression.Then the texture feature model and combined model based IVIM-DWI and texture feature parameters were established.The receiver operating characteristic（ROC）curves were used to evaluate the performances of IVIM-DWI,texture feature parameters,texture feature model and combined model in predicting lymph node metastasis from rectal adenocarcinoma.Results:In IVIM-DWI parameters,the D^*and f values of primary rectal adenocarcinoma were significantly different between the lymph node metastasis group and the non-lymph node metastasis group（Z=3.39,P=0.001,Z=-3.06,P=0.002）,the ADC and D had no statistical significance（both P>0.05）.A total of 828 texture feature parameters were obtained based on T₂WI of primary lesion of rectal adenocarcinoma,and 3 optimal texture feature parameters were selected,including firstorder＿Skewness,shape＿Sphericity and glcm＿Idn.The ROC curve results showed that the area under the curve（AUC）of D^*and f were 0.689 and 0.670,respectively.The AUC of 3 texture feature parameters were 0.651,0.628,0.631,respectively.The AUC of texture feature model was 0.775 and of the combined model was 0.803.Conclusion:IVIM-DWI parameters combined with T₂WI texture feature parameters in primary lesion of rectal adenocarcinoma can achieve good diagnostic efficacy for preoperative prediction of lymph node metastases with short diameter≤9 mm.Objective:To investigate the value of intravoxel incoherent motion diffusion weighted imaging（IVIM-DWI）and radiomics features of primary lesions of rectal adenocarcinoma in predicting postoperative recurrence.Methods: 89 patients with rectal adenocarcinoma confirmed by surgical pathology in our hospital were retrospectively analyzed from January 2016 to December 2019.All patients received preoperative rectum conventional MRI and IVIM-DWI scan,received radical resection of rectal cancer in our hospital.MRI data and follow-up data were collected and divided into recurrence group and non-recurrence group.The parameters（ADC,D,D*,f）of IVIM-DWI in the primary lesions of rectal adenocarcinoma were measured on GE AW4.5 workstation.Regions of interest（ROIs）were delineated layer by layer on axial T2 WI,IVIM-DWI（b=1200 s/mm2）and LAVA-Flex enhanced delayed phase images.GE Analysis Kit（AK）software was imported to extract 1037 3D radiomics features for each sequence.Statistically significant parameters were analyzed by receiver operating characteristic（ROC）curve.Four radiomics prediction models were established: Rad-score 1（Radiomics scores 1）based on T2 WI,Rad-Score 2 of IVIM-DWI,Rad-Score 3 of enhanced imaging,and Rad-Score 4 combined with all three.Then,multivariate logistic regression analysis was used to construct a nomogram combining Rad-Score 4 with independent risk factors.Area under the curve（AUC）was used to evaluate the efficacy of IVIM-DWI parameters,radiomics model,and combined model in predicting recurrence of rectal adenocarcinoma.The P-value of Hosmer-Lemeshow test,Brier value and consistency index（C-index）were calculated to assess the stability of the nomogram model.Finally,the model was internally validated by 200 times ten-fold cross-validation.Results: There were 21 cases of recurrence and 68 cases of non-recurrence.In IVIM-DWI parameters,the f value of the recurrence group was significantly higher than that of the non-recurrence group（P<0.05,AUC=0.808）.The AUC values of Rad-score1,Rad-Score 2,Rad-Score 3 and Rad-Score 4 were 0.817,0.852,0.843 and 0.932,respectively.The nomogram model,including Rad-Score 4 and f,AUC value was0.949.The average AUC,Brier value and C-index by 200 times ten-fold cross-validation were 0.944,0.052 and 0.949,respectively,indicating stability of the nomogram.Conclusion: IVIM-DWI parameters combined with multi-sequence and 3D MRI radiomics features of primary lesions are effective in predicting postoperative recurrence of rectal adenocarcinoma.