医学
接收机工作特性
曲线下面积
膀胱癌
生物标志物
病态的
决策树
放射科
癌症
核医学
泌尿科
内科学
人工智能
生物化学
化学
计算机科学
作者
B. Cao,Qing Li,Peng Xu,Kun Zhang,Shangli Cai,Shengxiang Rao,Ming Zhang,Yongming Dai,Shuai Jiang,Jianjun Zhou
标识
DOI:10.1016/j.crad.2024.01.031
摘要
To investigate whether the Vesical Imaging-Reporting and Data System (VI-RADS) could be used to develop a new non-invasive preoperative grade-prediction system to partially predict high-grade bladder cancer (HG-BC).The present study enrolled 89 primary BC patients prospectively from March 2022 to June 2023. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of VI-RADS for predicting HG-BC and muscle-invasive bladder cancer (MIBC) in the entire group. In the low VI-RADS (≤2) group, the decision tree-based method was used to obtain significant predictors and construct the decision-tree model (DT model). The performance of the DT model and low VI-RADS scores for predicting HG-BC was determined using ROC, calibration, and decision curve analyses.At a cut-off of ≥3, the specificity and positive predictive value of VI-RADS for predicting HG-BC in the entire group was 100%, and the area under the ROC curve (AUC) was 0.697. Among 65 patients with low VI-RADS scores, the DT model showed an AUC of 0.884 in predicting HG-BC compared to 0.506 for low VI-RADS scores. Calibration and decision curve analyses showed that the DT model performed better than the low VI-RADS scores.Most VI-RADS scores ≥3 correspond to HG-BCs. VI-RADS could be used as a grouping imaging biomarker for a pathological grade-prediction procedure, which in combination with the DT model for low VI-RADS (≤2) populations, would provide a potential preoperative non-invasive method of predicting HG-BC.
科研通智能强力驱动
Strongly Powered by AbleSci AI