联合临床、影像及bpMRI影像组学特征构建列线图模型预测高级别前列腺癌

打开文本图片集

关键词：前列腺肿瘤；双参数磁共振；影像组学；Gleason分级；列线图

中图分类号：R445.2；R737.25文献标志码：A 文章编号：1001-5779（2025）09-0873-08

DOI：10.3969/j.issn.1001-5779.2025.09.010

Abstract：Objective：To explore the valueof acombined nomogram model based onclinical data，imaging，and biparametric magnetic resonanceimaging（bpMRI）radiomics to predict high gradeof prostatecancer.Methods：A total of 137 prostatecancerpatientsadmited toourhospitalfromJanuary2O22 toDecember2O23 wereincludedinthis study. Theywere divided into a low-grade group（Gleason ⩽3+4 ）and a high-grade group（Gleason ≥ 4+3）according to pathological results. Also theywere randomlydivided intoa training group（ n=102 ）anda testing group（n=35）inratio of 3：1.The radiomics features of T2WI+ADC images were extractedand screened.Aradiomics model was established and radiomicsscores（Rad-Score）values werecalculated.The training group samples were used to analyze theclinical， imaging，andradiomicsparametersbyunivariateandmultivariateanalysistoscreentheindependentrisk factorsfor predicting high-grade prostatecancer.Thelogistic modelsofclinical imaging，radiomics，andcombined parameters were established and the nomogram was drawn.Delong test and receiver operating characteristic curve（ROC）were used to compare thediferencesand performanceamong models，andthedecisioncurveanalysiswasused toevaluate the clinical benefitsof each model.Results：Sixradiomics features related totheGleason gradeof prostatecancer were ultimatelyselected，aRad-Scorepredictionmodelwasestablished，andRad-Score werecalculated.Throughunivariate and multivariateanalyses，MRI-defined prostate specificantigen density（mPSAD），PI-RADSscore，andRad-Score wereincorporated intoanomogrammodel.TheAUCsof thismodel in the trainingandtest setswereO.881and 0.853，respectively，exceeding thoseof the clinical imaging model（AUC：0.793，O.755）and theRad-Score model （AUC：O.808，O.788），with statistically significant differences （ P<0.05 ）.Decision curve analysis demonstrated that the nomogram model provided greater clinical benefitcompared to theother models.Conclusion：The nomogram model based on clinical，imaging and T2WI + ADC imaging demonstrates favorable predictive performance for high-grade prostate cancer and holds potential for clinical application.

KeyWords： Prostatic cancer；Biparametric magnetic resonance imaging；Radiomics；Gleason grade；Nomogram

前列腺癌（Prostatecancer，PCa）作为常见的男性生殖系统恶性肿瘤，目前在我国已位居男性癌症发病率的第6位[。（剩余10766字）