基于MRI的可解释深度学习影像组学预测中轴型脊柱关节炎治疗反应
本研究为前瞻性研究，纳入183例拟接受TNFi治疗的中轴型脊柱关节炎患者，基于治疗前骶髂关节MRI提取深度学习与影像组学特征，构建DLRC模型预测ASAS40与ASAS20应答。独立测试集AUC分别为0.876和0.886，校准曲线与决策曲线分析支持其临床效用，较临床模型IDI显著改善。
要点
- 治疗前骶髂关节MRI可无创预测TNFi应答，辅助个体化分层。
- DLRC模型预测ASAS40的AUC为0.876，ASAS20为0.886。
- 联合临床数据较单纯临床模型IDI显著提升，支持多参数整合。
核心结论
- 基于MRI的可解释深度学习影像组学模型能准确预测axSpA患者TNFi治疗反应，提供无创的个体化治疗决策工具。
研究设计与样本
- 前瞻性研究，纳入183例确诊axSpA且TNFi治疗前接受骶髂关节MRI的患者，按约4:1分为训练集与测试集。
- 从MRI提取深度学习与影像组学特征，经可重复性分析、单变量筛选、冗余去除和LASSO回归四步筛选稳健预测因子。
- 以ASAS40和ASAS20为终点，将筛选特征与临床数据整合，采用L2正则化逻辑回归构建DLRC模型。
诊断标准与关键数值
- DLRC模型在独立测试集预测ASAS40的AUC为0.876（95%CI: 0.738-0.973）。
- DLRC模型在独立测试集预测ASAS20的AUC为0.886（95%CI: 0.753-0.978）。
- 较临床-only模型，DLRC模型预测ASAS40和ASAS20的IDI分别为0.234和0.283（均p<0.05）。
临床应用与阅片要点
- 阅片时可关注骶髂关节MRI的深度学习与影像组学特征，作为TNFi应答预测的辅助信息。
- 模型输出需结合临床数据综合解读，独立测试集校准曲线示良好一致性，决策曲线支持临床效用。
局限性
- 单中心前瞻性设计，样本量183例，外部泛化性需多中心验证。
- 深度学习与影像组学特征的可解释性及不同MRI设备/序列的稳健性未在原文中充分说明。
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
BACKGROUND: The clinical efficacy of tumor necrosis factor inhibitors (TNFi) in axial spondyloarthritis (axSpA) is limited by high non-response rates, necessitating accurate pre-treatment stratification. This study aimed to develop and validate MRI-based interpretable deep learning radiomics (DLR) models to predict treatment response to TNFi in patients with axSpA.
METHODS: In this prospective study, patients diagnosed with axSpA who underwent sacroiliac joint MRI before TNFi initiation were enrolled. Patients were allocated to training and test sets at an approximate 4:1 ratio. The endpoints were major (ASAS40) and moderate (ASAS20) improvement, as defined by the Assessment of SpondyloArthritis International Society criteria. Deep learning and radiomics features were extracted from MRI. A four-step selection process, including reproducibility analysis, univariable filtering, redundancy reduction, and LASSO regression, was used to distill robust predictors. These were then integrated with clinical data to build DLR-clinical (DLRC) models using L2-regularized logistic regression.
RESULTS: A total of 183 patients (mean age, 26.5 ± 9.0 years; 121 [66.1%] males) were analyzed. The DLRC models were validated on the independent test set, with high area under the receiver operating characteristic curve (AUC) values for predicting ASAS40 (0.876, 95%CI: 0.738-0.973) and ASAS20 (0.886, 95%CI: 0.753-0.978) responses. Calibration curves indicated good model agreement, and decision curve analysis (DCA) confirmed their clinical utility. Compared to the clinical-only model, the DLRC models showed significant integrated discrimination improvement (IDI) for ASAS40 and ASAS20 (0.234 and 0.283, all p < 0.05) prediction, respectively.
CONCLUSION: MRI-based interpretable DLR models accurately predicted TNFi treatment response in axSpA patients, offering a non-invasive tool to guide personalized therapy.
引用
- Xie Z, Lu K, Liu Y, et al. MRI-based interpretable deep learning radiomics for predicting treatment response in axial spondyloarthritis. Eur J Radiol. 2026. PMID: 42085769. doi:10.1016/j.ejrad.2026.112903
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。