基于MRI的多参数影像组学构建预测前列腺癌盆腔淋巴结转移风险预警模型OACSTPCD

Objective:To analyze the value of MRI multiparameter imaging in predicting pelvic lymph node metastasis in prostate cancer(PCa).Methods:A total of 131 pathologically confirmed PCa patients admitted to our hospital from March 2018 to March 2023 were selected and divided into training group and validation group.Dynamic contrast enhanced MRI(DCE-MRI),prostate magnetic resonance imaging(MRI)plain scan and diffusion-weighted imaging(DWI)were performed before surgery.A total of 91 patients in the training group were divided into metastatic group(n=34)and non-metastatic group(n=57)according to whether they had pelvic lymph node metastasis.Clinical and imaging parameters of patients were collected and univariate and multivariate Logistic regression analyses were performed to determine independent predictors.ROC curve was used to evaluate the efficacy of the prediction model.Results:Univariate analysis showed that PSA,PI-RADS score,Gleason score,ADC value,lymph signal and DCE image were significantly different between the two groups(P＜0.05).Multivariate analysis determined that ADC value,lymphatic signal,DCE image,PI-RADS score and Gleason score were independent predictors(OR values were 0.014,9.507,21.646,8.965,4.314,respectively,all P＜0.05).The area under ROC curve of the established model in the training group was 0.936(95％CI:0.890-0.982),the sensitivity was 97.1％,the specificity was 77.8％,and the accuracy was 87.5％.In the validation group,the area was 0.929(95％CI:0.867-0.990),the sensitivity was 88.2％,the specificity was 93.0％,and the accuracy was 90.6％.Conclusion:ADC value,lymphatic signal,DCE image,PI-RADS score and Gleason score are independent factors for predicting pelvic lymph node metastasis in PCa patients,and the risk warning model established based on these factors has good predictive efficacy.