深度学习实现类风湿关节炎关节损伤进展的自动检测——模型开发与外部验证
AuRA算法基于两项既往临床研究(n=367)的手足X线片训练,用于预测Sharp-van der Heijde总分,并在图尔库大学医院数据(n=205)中外部验证,其均方根误差为23.6,优于两种表现最佳的RA2-DREAM算法(均方根误差35.0和35.6)。在54例患者平均间隔4.6年的随访中,算法预测的纵向变化与专家评估评分变化显著相关(Pearson R=0.74,P<0.001)。
要点
- AuRA外部验证均方根误差23.6,优于两种RA2-DREAM算法(35.0和35.6)。
- 算法预测的纵向评分变化与专家评估变化显著相关(Pearson R=0.74,P<0.001)。
- 算法已公开于Docker平台,可用于自动检测影像学进展以减少人工评分负担。
核心结论
- AuRA在外部验证中表现最佳,并显示出检测关节损伤纵向变化的潜力。
- 该算法可便捷地用于未来自动检测影像学进展,减少繁琐的人工评分需求。
研究设计与样本
- 算法在类风湿关节炎逆向工程评估与方法学对话(RA2-DREAM)挑战赛中开发,训练数据来自两项既往临床研究(n=367)的手足X线片,用于预测专家制定的Sharp-van der Heijde总分。
- 外部验证使用图尔库大学医院数据(n=205),并与两种表现最佳的RA2-DREAM算法比较。
- 另从54例患者的另一次对照就诊中提取额外X线片组,平均时间间隔4.6年。
主要发现
- 外部验证队列中AuRA的均方根误差为23.6,两种表现最佳的RA2-DREAM算法均方根误差分别为35.0和35.6。
- 性能提升主要归因于专家评估评分较高时误差较低。
- 算法预测的纵向变化与专家评估评分变化显著相关(Pearson R=0.74,P<0.001)。
临床怎么用
- 原文未据此给出推荐。
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)。
原文
OBJECTIVES: Although deep learning has demonstrated substantial potential in automatic quantification of joint damage in RA, evidence for detecting longitudinal changes at an individual patient level is lacking. Here, we introduce and externally validate our automated RA scoring algorithm (AuRA), and demonstrate its utility for monitoring radiographic progression in a real-world setting. METHODS: The algorithm, originally developed during the Rheumatoid Arthritis 2-Dialogue for Reverse Engineering Assessment and Methods (RA2-DREAM) challenge, was trained to predict expert-curated Sharp-van der Heijde total scores in hand and foot radiographs from two previous clinical studies (n = 367). We externally validated AuRA against data (n = 205) from Turku University Hospital and compared the performance against two top-performing RA2-DREAM solutions. Finally, for 54 patients, we extracted additional radiograph sets from another control visit to the clinic (average time interval of 4.6 years). RESULTS: In the external validation cohort, with a root mean square error (RMSE) of 23.6, AuRA outperformed both top-performing RA2-DREAM algorithms (RMSEs 35.0 and 35.6). The improved performance was explained mostly by lower errors at higher expert-assessed scores. The longitudinal changes predicted by our algorithm were significantly correlated with changes in expert-assessed scores (Pearson's R = 0.74, P < 0.001). CONCLUSION: AuRA had the best external validation performance and demonstrated potential for detecting longitudinal changes in joint damage. Available from https://hub.docker.com/r/elolab/aura, our algorithm can easily be applied for automatic detection of radiographic progression in the future, reducing the need for laborious manual scoring.
引用
- Venäläinen MS, Biehl A, Holstila M, et al. Deep learning enables automatic detection of joint damage progression in rheumatoid arthritis-model development and external validation. Rheumatology (Oxford). 2025. PMID: 38597875. doi:10.1093/rheumatology/keae215
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。