基于X线与CT多模态融合的深度学习模型预测急性椎体骨折
本研究开发并验证了一种基于X线与CT的深度学习模型,用于诊断急性椎体骨折,并与脊柱外科医生进行诊断准确性比较。输入级融合模型性能最佳,椎体水平F1分数达0.953,其敏感度(0.962)与特异度(0.981)均优于高年资脊柱外科医生(0.925、0.955)。
要点
- X线与CT输入级融合的YOLOv11模型对MRI定义的急性椎体骨折诊断性能最佳。
- 阈值0.24时召回率达1.0,适合作为减少漏诊的筛查阈值。
- 模型敏感度与特异度均高于高年资脊柱外科医生。
核心结论
- 基于YOLOv11的输入级融合模型能利用常规X线与CT准确预测MRI定义的急性椎体骨折。
研究设计与样本
- 单中心回顾性诊断准确性研究。
- 纳入2024年1月至2025年6月于南京医科大学附属江宁医院脊柱外科住院并诊断为椎体压缩骨折患者的X线、CT及MRI图像。
- 使用3D Slicer的「Landmark Registration」工具实现脊柱图像配准。
- 采用YOLOv11目标检测技术构建深度学习模型,以MRI作为定义急性椎体骨折的参考标准。
诊断标准与关键数值
- 输入级融合模型F1分数为0.953,精度-召回曲线下面积为0.990。
- 选择阈值0.24时模型召回率为1.0,精确度为0.855。
- 模型敏感度为0.962,高于高年资脊柱外科医生的0.925。
- 模型特异度为0.981,高于高年资脊柱外科医生的0.955。
- 输入级融合配置性能优于仅X线、仅CT及决策级融合配置。
临床应用与阅片要点
- 输入级融合模型可作为急性椎体骨折的筛查工具,尤其适用于无法立即行MRI的情况。
- 阈值0.24对应召回率1.0,适合临床应用中以减少漏诊为首要目标的场景。
- 模型性能优于高年资脊柱外科医生,提示具有辅助减少漏诊与误诊的潜力。
局限性
- 单中心回顾性研究设计限制了结果的普遍适用性。
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
RATIONALE AND OBJECTIVES: The study aimed to develop and validate a deep learning (DL) model based on X-ray and computed tomography (CT) to diagnose acute vertebral fractures (VFs), and to compare its diagnostic accuracy against spine surgeons.
MATERIALS AND METHODS: This single-center, retrospective diagnostic accuracy study mainly gathered X-ray, CT, and magnetic resonance imaging (MRI) images of patients diagnosed with vertebral compression fractures who were hospitalized in the Department of Spinal Surgery of the Affiliated Jiangning Hospital of Nanjing Medical University from January 2024 to June 2025. Spine image registration was achieved using the "Landmark Registration" tool in 3D Slicer. The latest You-Only-Look-Once (YOLO) v11 target detection technology was employed to construct the DL model, which was trained using multimodal images. Four configurations (X-ray only, CT only, input-level fusion, and decision-level fusion) were evaluated using metrics such as mean average precision, precision, recall, and F1 score. Finally, the model's performance was compared against that of middle-grade and senior spine surgeons.
RESULTS: The input-level fusion model achieved the best performance, with an area under the precision-recall curve of 0.990 and an F1-score of 0.953 at the vertebra level. To minimize missed diagnoses, a threshold of 0.24 was selected, achieving a recall of 1.0 and a precision of 0.855. The model significantly outperformed middle-grade surgeons and demonstrated superior sensitivity (0.962 vs. 0.925) and specificity (0.981 vs. 0.955) compared to senior surgeons.
CONCLUSION: The YOLOv11-based input-level fusion model accurately predicts MRI-defined acute VFs using routine X-ray and CT imaging.
引用
- Chen, Hu, Wang, et al. Deep Learning Method Based on Multimodal Medical Image Fusion for Predicting Acute Vertebral Fracture. Acad Radiol. 2026. PMID: 42692873. doi:10.1016/j.acra.2026.08.076
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。