基于CT的多模态融合模型预测钝性胸部创伤结局:一项多中心研究
该研究回顾性分析三个医疗中心共337例钝性胸部创伤患者,提取临床变量、影像组学特征、深度学习特征及代表时间变化的delta特征,经LASSO回归筛选后构建临床模型、影像模型和融合模型。肋骨骨折数、多发伤和血气胸-肺比(HPR)为独立预后因素。融合模型(尤其delta-clinical-DLR模型)在训练集和测试集的AUC分别为0.85(95%CI 0.80-0.90)和0.86(95%CI 0.77-0.95)。与单纯临床模型相比,融合模型显著改善净重分类(NRI最高0.76)和综合判别(IDI最高0.26)。作者认为,整合多时间点CT与临床变量的多模态融合模型可显著提升钝性胸部创伤的预后预测性能,为个体化风险预测和临床决策提供工具。
要点
- 在337例钝性胸部创伤患者中,肋骨骨折数、多发伤和血气胸-肺比(HPR)被识别为独立预后因素。
- delta-clinical-DLR融合模型在训练集AUC为0.85(95%CI 0.80-0.90),在测试集AUC为0.86(95%CI 0.77-0.95)。
- 与单纯临床模型相比,融合模型的净重分类改善(NRI)最高达0.76,综合判别改善(IDI)最高达0.26。
- 该模型整合了临床变量、影像组学特征、深度学习特征及代表时间变化的delta特征,并纳入急诊与随访CT扫描。
- 作者认为该多模态融合模型可显著提升钝性胸部创伤的预后预测性能,为个体化风险预测和临床决策提供工具。
适用人群与场景
- 回顾性纳入三个医疗中心的337例钝性胸部创伤(BCT)患者。
- 获取临床资料和CT影像数据,包括急诊和随访扫描。
研究设计与关键结果
- 研究类型:回顾性多中心研究,共三个医疗中心,样本量337例钝性胸部创伤患者。
- 独立预后因素:肋骨骨折数、多发伤、血气胸-肺比(HPR)。
- 融合模型(尤其delta-clinical-DLR模型)训练集AUC 0.85(95%CI 0.80-0.90),测试集AUC 0.86(95%CI 0.77-0.95)。
- 与单纯临床模型相比,净重分类改善(NRI)最高0.76,综合判别改善(IDI)最高0.26。
- 模型开发包括单纯临床模型、单纯影像模型和融合模型,采用LASSO回归筛选特征,以AUC、校准曲线和决策曲线分析评价性能。
证据类型、等级与局限
- 证据类型:回顾性多中心研究。
- 局限:原文未报告局限性。
对床旁实践的改变
- 作者认为,整合多时间点CT影像与临床变量的多模态融合模型可显著提升钝性胸部创伤的预后预测性能,为个体化风险预测和临床决策提供工具。
提炼说明
- 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开
BACKGROUND: This study aimed to develop a multimodal predictive model that integrates clinical variables, radiomic features (RFs), and deep learning-based features (DLFs) to improve prognostic accuracy in patients with blunt chest trauma (BCT).
METHODS: We retrospectively analyzed 337 patients with BCT from three medical centers. Clinical and CT imaging data, including emergency and follow-up scans, were obtained. Features including radiomic (RF) and deep learning-based (DLF) descriptors along with delta features representing temporal changes were extracted. After sequential feature selection, the least absolute shrinkage and selection operator (LASSO) regression was applied to identify optimal features. Model development included clinical-only, imaging-only, and fused models, with performance evaluated using AUC, calibration curves, and decision curve analysis.
RESULTS: Rib fracture count, multiple injuries, and hemopneumothorax-to-lung ratio (HPR) were identified as independent prognostic factors. The fusion model, particularly the delta-clinical-DLR model, achieved AUCs of 0.85 (95 % CI: 0.80-0.90) and 0.86 (95 % CI: 0.77-0.95) in the training and test sets, respectively. Significant improvements in net reclassification (NRI up to 0.76) and integrated discrimination (IDI up to 0.26) were observed compared to the clinical models alone.
CONCLUSIONS: Integrating multi-timepoint CT imaging with clinical variables through a multimodal fusion model significantly enhances the prognostic performance of BCT, providing a robust tool for individualized risk prediction and clinical decision-making.
参考来源
- Zhao T, Li D, Wu M, et al. A CT-based multimodal fusion model for predicting outcomes in blunt chest trauma: A multicenter study. Am J Emerg Med. 2026. PMID: 41576754. doi:10.1016/j.ajem.2025.12.038
供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。