Study On Predictive Diagnosis Model Of Prostate Cancer Based On Clinical Data

Objective Prostate cancer is one of the common diseases that threaten the health of men.Currently,PSA is mainly used for diagnosis in clinical practice,but there are still many shortcomings.They are susceptible to many factors such as the patient's own physique(prostate volume,total human blood circulation)and other factors.If this effect can be eliminated,it will improve the accuracy of early diagnosis of prostate cancer and reduce the rate of missed diagnosis.This study uses CRT classification trees and Logistic regression analysis to expeditiously construct a diagnostic model for obesity and prostate cancer,and applies ROC curves to evaluate their effectiveness,providing a basis for prostate cancer diagnosis.Methods A total of 370 cases of prostatosis cases were collected from the Second Affiliated Hospital of Soochow University.All cases were tested for PSA and prostate biopsy,of which 152 were prostate cancer and 218 were non-prostate cancer.The clinical data of 370 patients such as age,prostate volume(PV),PSA,height,weight,BMI,body surface area(BSA),PSAM and PSAMR were analyzed by CRT classification tree and logistic regression analysis.Screening with prostate Indexes related to cancer diagnosis,and construct a prostate cancer diagnosis model.Finally,the ROC curve is used to verify the accuracy of the model.Results Independent sample T test results showed that age?PV?PSA?PSAM and PSAMR had significant statistical differences between the prostate cancer group and the non-prostate cancer group,with a P value of less than 0.05;Pearson correlation analysis showed that no single clinical indicator and prostate cancer diagnosis were significant.Correlation,did not meet the clinical indicators of R>0.5 and P value<0.05;CRT classification tree analysis showed that the importance of standardization of PSAMR,age,PSA,PV,and PSAM was 100%,42.5%,39.4%,39.0%,and 36.9%.ROC AUC is 0.878,sensitivity is 75.66%,specificity is 90.8%,accuracy is 84.59,and it has high reliability;Logistic regression analysis shows that age,BSA,PSA,PSAM,and PSAMR are all independent diagnosis of prostate cancer Logistic(p)=11.995+0.055*age+3.954*BSA+0.323*PSA-0.118*PSAM+1.619*PSAMR,ROC AUC is 0.879,sensitivity is 75.50%,specificity It is 91.70%,the accuracy is 85.10%,and it has high reliability.Conclusion Age,PSA,PSAM,PSAMR,PV,and BSA have significant effects on the diagnosis of prostate cancer.Combined application of clinical characteristics of patients can help to improve the diagnosis model of prostate cancer,improve the accuracy of diagnosis,reduce the rate of misdiagnosis,and avoid the impact of meaningless puncture biopsies on patients' bodies.