基于影像组学的IVIM-DWI用于肾移植功能障碍的早期无创评估
本回顾性研究纳入97例肾移植受者,以eGFR 60 ml/min/1.73 m2为界分为肾功能正常组与受损组,利用术后3T IVIM-DWI生成的参数图提取影像组学特征,比较单b值DWI与IVIM参数图模型对移植肾功能受损的预测效能,发现联合多b值DWI与IVIM图可提升诊断准确性。
要点
- 以eGFR 60 ml/min/1.73 m2为界定义的移植肾功能受损,IVIM参数图影像组学模型验证AUC为0.790(95%CI 0.607-0.937)。
- 联合多b值DWI与IVIM图的整合模型验证AUC为0.790(95%CI 0.600-0.951),训练AUC为0.816。
- Dslow图的中位直方图强度是预测移植肾功能受损最有影响力的特征。
核心结论
- 基于IVIM-DWI的影像组学模型(含单b值DWI与IVIM参数图)可用于肾移植受者移植肾功能受损的早期无创评估。
- 联合多b值DWI与IVIM成像可明显提高诊断准确性,构成一个稳健的无创早期检测框架。
- 本研究首次全面比较了基于IVIM-DWI的影像组学模型(包括单b值DWI和IVIM参数图)对移植肾功能障碍早期评估的预测效能。
研究设计与样本
- 回顾性研究,纳入97例肾移植受者,平均年龄36.77 ± 10.71岁。
- 按eGFR 60 ml/min/1.73 m2为界分为肾功能正常组与受损组。
- 患者随机分为训练组(n = 68)与验证组(n = 29)。
- 术后采用3T扫描仪行11个b值的IVIM-DWI MRI,生成ADC、Dslow、Dfast、PF参数图。
- 采用全移植肾3D手动分割,每个数据集提取1604个影像组学特征,经ANOVA、Relief和RFE筛选后以10种机器学习算法分类。
诊断标准与关键数值
- 基于IVIM参数图(ADC、Dslow、Dfast、PF)的影像组学模型:ANOVA筛选特征联合自编码器(AE)分类,验证AUC 0.790(95%CI 0.607-0.937),训练AUC 0.770。
- 联合多b值DWI与IVIM图的整合模型:ANOVA筛选16个特征联合朴素贝叶斯(NB)分类,验证AUC 0.790(95%CI 0.600-0.951),训练AUC 0.816。
- AE与NB分类器在所有模型配置中均表现出最强的判别性能。
- Dslow图的中位直方图强度是预测移植肾功能受损最有影响力的特征。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
OBJECTIVE: To develop and rigorously validate radiomics-based predictive models using postoperative intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) MRI for the early, noninvasive assessment of impaired renal allograft function (IRF) in kidney transplant recipients.
METHODS: This retrospective study included 97 kidney transplant recipients (mean age, 36.77 ± 10.71 years), categorized into normal or impaired renal function groups based on an estimated glomerular filtration rate (eGFR) cutoff of 60 ml/min/1.73 m2. Patients were randomly assigned to training (n = 68) or validation (n = 29) groups. Postoperative IVIM-DWI MRI with 11 b-values was performed on a 3T scanner, generating parametric maps (apparent diffusion coefficient (ADC), slow diffusion coefficient (Dslow), fast diffusion coefficient (Dfast), perfusion fraction (PF)). Whole-graft 3D manual segmentation was used to extract 1604 radiomic features per dataset. Feature selection was performed through analysis of variance (ANOVA), Relief, and recursive feature elimination (RFE), followed by classification using ten machine learning algorithms, including auto-encoder (AE) and naïve Bayes (NB). Performance was evaluated using receiver operating characteristic (ROC) analysis, with area under the curve (AUC), accuracy, sensitivity, and specificity as metrics.
RESULTS: Radiomics models based on IVIM-derived parametric maps (ADC, Dslow, Dfast, PF) achieved superior diagnostic performance, with a validation AUC of 0.790 (95% confidence interval (CI) 0.607-0.937) using ANOVA-based feature selection and AE classification, and a training AUC of 0.770. Integrative models combining multi-b-value DWI and IVIM maps further enhanced predictive power, achieving a validation AUC of 0.790 (95% CI 0.600-0.951) and a training AUC of 0.816, utilizing 16 features selected via ANOVA and classified with the NB algorithm. AE and NB classifiers consistently exhibited the strongest discriminative performance across all model configurations. Notably, the median histogram intensity from the Dslow map was the most influential feature for predicting impaired renal function.
CONCLUSIONS: This study is the first to comprehensively compare the predictive performance of radiomics models based on IVIM-DWI, including both single b-value DWI and IVIM parametric maps, for early assessment of renal allograft dysfunction. The integrative use of multi-b-value DWI and IVIM imaging markedly improves diagnostic accuracy, demonstrating a robust noninvasive framework for early detection of renal allograft dysfunction.
引用
- Liu X, Chen L, Yang L, et al. Radiomics-based IVIM-DWI for early noninvasive assessment of renal allograft dysfunction. Abdom Radiol (NY). 2025. PMID: 41081883. doi:10.1007/s00261-025-05226-x
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。