IgeaNet:利用胸片进行机会性骨质疏松筛查的深度学习模型
本研究开发并内部验证了一种多模态深度学习模型IgeaNet,整合胸片与年龄、性别、身高、体重等临床元数据对骨密度进行分类。模型在内部测试集上总体准确率为85.8%(95%CI 78.5-90.9%),AUC-macro为0.95。研究认为该方法有望在不增加辐射暴露的情况下对骨密度进行初步机会性分类,但结果来自内部验证,临床应用前需外部前瞻性研究证实。
要点
- 胸片结合临床元数据可对骨密度进行机会性初步分类,无需额外辐射暴露。
- 骨质疏松类别的敏感度为82.3%、特异度为91.9%,适合作为筛查初筛工具。
- 分类结果为初步筛查性质,不能替代DEXA,临床决策前仍需标准骨密度检查。
核心结论
- 整合胸片与临床数据的残差网络结合注意力机制方法,可在不增加辐射暴露的情况下对骨密度进行机会性初步分类。
- 研究结果具有假设生成性质,仅来自内部验证,临床实施前需要外部前瞻性研究确认。
研究设计与样本
- 单中心回顾性研究,纳入2018年1月至2024年12月期间833例接受胸片和双能X线吸收法检查的成年患者。
- 胸片与DEXA检查间隔在12个月内。
- 数据集划分为训练集589例、验证集124例、测试集120例。
- 采用卷积神经网络分支处理胸片,密集分支处理患者人口学资料(年龄、性别、身高、体重),两分支特征融合后进入全连接分类器端到端优化。
关键影像表现/征象
- DEXA解剖部位分布显示腕部为决定性部位者302例(36.3%),髋部384例(46.1%),腰椎141例(16.9%)。
- 纳入腕部测量结果改变了115例(13.8%)患者的诊断分类,其中34例由正常改为骨量减少,14例由正常改为骨质疏松,67例由骨量减少改为骨质疏松。
- 腕部皮质骨可能揭示在以松质骨为主的部位未被检出的骨密度降低,体现了不同部位骨成分差异带来的不同诊断效能。
诊断标准与关键数值
- 内部测试集模型总体准确率为85.8%(95%CI:78.5-90.9%),F1-macro为0.86,AUC-macro为0.95。
- 类别特异性能:正常类敏感度84.2%、特异度91.4%;骨量减少类敏感度90.2%、特异度77.1%;骨质疏松类敏感度82.3%、特异度91.9%。
- 总体误分类率为14.2%。
临床应用与阅片要点
- 该模型可作为一种低成本的骨密度机会性筛查工具,适用于已行胸片检查而未经DEXA筛查的人群。
- 重点关注模型对骨量减少类特异度较低(77.1%)的特点,该类别误判可能导致假阳性增加,需结合临床随访确认。
- 阅片时若对老年患者胸片具备骨小梁稀疏等间接表现,可提示进一步DEXA检查,但不应仅凭胸片下骨质疏松诊断。
局限性
- 单中心回顾性设计,样本量有限(833例),结果仅经内部验证。
- 研究结果具有假设生成性质,缺乏外部前瞻性多中心验证。
- 胸片与DEXA检查间隔最长可达12个月,期间骨密度变化可能影响分类标签的准确性。
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
PURPOSE: To develop and validate a multi-modal deep learning model for opportunistic screening of osteoporosis by classifying bone mineral density (BMD) from chest radiographs integrated with clinical metadata.
MATERIALS AND METHODS: This single-centre retrospective study included 833 adult patients who underwent chest radiography and dual-energy X-ray absorptiometry (DEXA) within 12 months (January 2018-December 2024). A convolutional neural network (CNN) model was developed using a convolutional branch to process chest radiographs, while a dense branch processed patient demographics (age, sex, height, and weight). Clinical data available in the DEXA report were also included. The two representations were merged into a fully connected classifier optimized end-to-end.
RESULTS: The dataset comprised training (n = 589), validation (n = 124), and testing (n = 120) sets. On the internal test set, the model achieved 85.8% accuracy (95% CI: 78.5-90.9%), F1-macro 0.86, and AUC-macro 0.95. Class-specific performance showed sensitivity/specificity of 84.2%/91.4% for normal, 90.2%/77.1% for osteopenia, and 82.3%/91.9% for osteoporosis, with 14.2% misclassification rate. Analysis of DEXA anatomical site distribution revealed wrist as the determining site in 302 patients (36.3%), hip in 384 patients (46.1%), and lumbar spine in 141 patients (16.9%). Inclusion of wrist measurements changed diagnostic classification in 115 patients (13.8%), with 34 patients reclassified from normal to osteopenia, 14 from normal to osteoporosis, and 67 from osteopenia to osteoporosis. This differential diagnostic yield reflects distinct bone composition differences, as wrist cortical bone may reveal mineral density reductions undetected at predominantly trabecular sites.
CONCLUSIONS: Integration of chest radiographs and clinical data by residual network with attention represents a feasible approach for the preliminary opportunistic classification of BMD without additional radiation exposure. However, these findings are hypothesis-generating and derive from internal validation; external prospective studies are required before clinical implementation.
引用
- Mattiussi, Camponovo, Magoga, et al. IgeaNet: a deep learning model for opportunistic osteoporosis screening from chest X-ray images. Radiol Med. 2026. PMID: 42753041. doi:10.1007/s11547-026-02277-y
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。