Application of MRI in Diagnosis and Prognosis Evaluation of Prostate Cancer
Objective:To analyze the technical points and clinical value of MRI imaging in diagnosis and prognosis evaluation of prostate cancer.Methods:A total of 120 patients with prostate cancer from January 2021 to December 2022 were selected.MRI imaging omics method and pathological examination were used to extract image texture features and construct models,and pathological examination results and main texture features were analyzed to evaluate the application value of the models.Results:Pathological examination showed 61 cases of prostate cancer and 59 cases of prostate hyperplasia.The image texture features showed 1129 images in 6 categories.Model analysis showed that the prostatic lesion prediction model had good application efficacy,with AUC 0.87,95％confidence interval(0.72-1.00),sensitivity 81.0％,specificity 90.0％,model accuracy 100.0％,positive predictive value 100.0％and negative predictive value 85.00％in the training group.In the verification group,AUC 0.98,95％confidence interval(0.93-1.00),sensitivity 85.0％,specificity 85.0％,model accuracy 85.0％,positive predictive value 85.0％,negative predictive value 85.0％.Conclusions:MRI imaging omics can assist physicians to effectively detect prostate lesions and promote lesion identification.Based on ADC images,imaging features can be determined and models built,which can more accurately distinguish prostate cancer and prostate hyperplasia and guide clinicians to treat diseases scientifically.
万方数据
维普
