| 摘要: |
| 目的 探究基于多参数磁共振成像(mpMRI)影像组学特征及临床特征在低、高级前列腺癌(PCa)Gleason术前评估中的临床应用价值。方法 选取2024年1月~2024年12月年来我院住院治疗且经病理证实为PCa的180例患者作为研究对象,根据Gleason分级结果,将PCa患者被分为两组:Gleason分级为1~2级的患者归为低级别PCa组,而Gleason分级为3~5级的患者归为高级别PCa组。收集患者临床资料,并以ROC曲线评估mpMRI影像组学特征及临床特征对低、高级前列腺癌(PCa)术前评估中的临床应用价值。以单因素及多因素Logistic回归分析术前评估高级别PCa的危险因素,并构建列线图模型。结果 高级别PCa组前列腺癌患者的mpMRI影像组学评分明显高于低级别PCa组(P＜0.05)。mpMRI影像组学评分＞0.53分、肿瘤直径、PSA、PSAD及N分期为术前评估为高级别PCa的独立危险因素(P<0.05)。基于上述危险因素构建预测术前评估为高级别PCa的列线图模型,C-index为0.994(95%CI:0.971-0.999)；校准曲线接近理想状态,显示了模型的高准确性。ROC曲线下的面积(AUC)达到0.996(95%CI:0.972~0.999),表明该模型在区分前列腺癌患者方面表现极佳。此外,在2%到100%的预测概率范围内,模型均能净获益。结论 基于mpMRI影像组学与临床特征构建预测术前评估为高级别PCa的列线图模型,具有较好的预测效能。 |
| 关键词: mpMRI影像组学 临床特征 前列腺癌 Gleason分级 |
| DOI： |
| 分类号: |
| 基金项目:徐州市卫生健康委员会科技项目，项目编号：XWKYSL20220243 |
|
| Preoperative Assessment of Prostate Cancer Gleason Grade Grouping Based on mpMRI Radiomics and Clinical Features |
|
宋腾腾
|
|
Suining Branch of Affiliated Hospital of Xuzhou Medical University
|
| Abstract: |
| Objective To explore the clinical application value of multiparametric magnetic resonance imaging (mpMRI) radiomics features and clinical characteristics in the preoperative assessment of low- and high-grade prostate cancer (PCa) based on Gleason grading. Methods A total of 180 patients diagnosed with PCa by pathology and hospitalized in our institution from January 2024 to December 2024 were selected as the study subjects. According to the Gleason grading results, PCa patients were divided into two groups: those with Gleason grades 1–2 were classified as the low-grade PCa group, while those with Gleason grades 3–5 were classified as the high-grade PCa group. Clinical data were collected, and the ROC curve was used to evaluate the clinical application value of mpMRI radiomics features and clinical characteristics in the preoperative assessment of low- and high-grade PCa. Univariate and multivariate logistic regression analyses were performed to identify risk factors for preoperative assessment of high-grade PCa, and a nomogram model was constructed. Results The mpMRI radiomics score was significantly higher in the high-grade PCa group compared to the low-grade PCa group (P < 0.05). An mpMRI radiomics score>0.53, tumor diameter, PSA, PSAD, and N stage were identified as independent risk factors for preoperative assessment of high-grade PCa (P <0.05). A nomogram model was constructed based on these risk factors to predict preoperative high-grade PCa, with a C-index of 0.994 (95% CI: 0.971–0.999). The calibration curve was close to the ideal state, demonstrating the high accuracy of the model. The area under the ROC curve (AUC) reached 0.996 (95%CI: 0.972–0.999), indicating excellent performance of the model in distinguishing PCa patients. Additionally, the model provided positive net benefits across a prediction probability range of 2% to 100%. Conclusion The nomogram model constructed based on mpMRI radiomics and clinical features for preoperative assessment of high-grade PCa demonstrates good predictive performance. |
| Key words: mpMRI radiomics clinical features prostate cancer Gleason grading |
