基于改良2算法定性标准适配的脊柱骨质疏松性压缩骨折X线深度学习分类的泛化性
本研究回顾性构建本地数据集(1790例受试者、15050个胸腰椎椎体),并采用MrOS研究数据集(6个临床中心5994名男性的胸腰椎X线片),训练五种深度学习算法对每个椎体进行分类;最佳集成平均模型在本地测试集和MrOS测试集AUC-ROC分别为0.948和0.936,以优先保证PPV的阈值下敏感度分别为54.5%和47.8%、特异度分别为99.7%和99.6%、PPV分别为89.8%和94.8%。
要点
- 最佳集成模型在本地与MrOS测试集AUC-ROC分别为0.948和0.936,均大于0.90。
- 高PPV阈值下敏感度仅54.5%和47.8%,特异度达99.7%和99.6%。
- 模型对真实临床数据集显示一定泛化性,可能适用于未来机会性骨质疏松筛查工具。
核心结论
- 模型在两个数据集上AUC-ROC均大于0.90,显示对真实临床数据集有一定泛化性,并具有作为未来机会性骨质疏松筛查工具的合适性能。
研究设计与样本
- 回顾性研究,本地数据集含1790例受试者、15050个胸腰椎椎体,每个椎体按改良2算法定性标准适配进行标注;MrOS研究数据集提供6个临床中心5994名男性的胸腰椎X线片;使用两个数据集训练五种深度学习算法对每个椎体进行分类。
诊断标准与关键数值
- 最佳集成平均模型在本地数据集测试集AUC-ROC为0.948、在MrOS数据集测试集AUC-ROC为0.936;将 cutoff 阈值设为优先保证PPV后,本地测试集敏感度54.5%、特异度99.7%、PPV 89.8%,MrOS测试集敏感度47.8%、特异度99.6%、PPV 94.8%。
临床应用与阅片要点
- 作者认为该模型对真实临床数据集显示一定泛化性,并具有作为未来机会性骨质疏松筛查工具的合适性能。
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
RATIONALE AND OBJECTIVES: Spinal osteoporotic compression fractures (OCFs) can be an early biomarker for osteoporosis but are often subtle, incidental, and underreported. To ensure early diagnosis and treatment of osteoporosis, we aimed to build a deep learning vertebral body classifier for OCFs as a critical component of our future automated opportunistic screening tool.
MATERIALS AND METHODS: We retrospectively assembled a local dataset, including 1790 subjects and 15,050 vertebral bodies (thoracic and lumbar). Each vertebral body was annotated using an adaption of the modified-2 algorithm-based qualitative criteria. The Osteoporotic Fractures in Men (MrOS) Study dataset provided thoracic and lumbar spine radiographs of 5994 men from six clinical centers. Using both datasets, five deep learning algorithms were trained to classify each individual vertebral body of the spine radiographs. Classification performance was compared for these models using multiple metrics, including the area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, and positive predictive value (PPV).
RESULTS: Our best model, built with ensemble averaging, achieved an AUC-ROC of 0.948 and 0.936 on the local dataset's test set and the MrOS dataset's test set, respectively. After setting the cutoff threshold to prioritize PPV, this model achieved a sensitivity of 54.5% and 47.8%, a specificity of 99.7% and 99.6%, and a PPV of 89.8% and 94.8%.
CONCLUSION: Our model achieved an AUC-ROC>0.90 on both datasets. This testing shows some generalizability to real-world clinical datasets and a suitable performance for a future opportunistic osteoporosis screening tool.
引用
- Dong Q, Luo G, Lane NE, et al. Generalizability of Deep Learning Classification of Spinal Osteoporotic Compression Fractures on Radiographs Using an Adaptation of the Modified-2 Algorithm-Based Qualitative Criteria. Acad Radiol. 2023. PMID: 37438161. doi:10.1016/j.acra.2023.04.023
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。