文献速递骨肌技术与安全DR

基于分层持续学习的多关节关节病分级分类模型

Classification models for arthropathy grades of multiple joints based on hierarchical continual learning.
PubMed 文献 · Radiol Med · 2025-03-24

本研究纳入来自三家三级医院的1371张膝、肘、踝、肩、髋关节X线片,开发分层持续学习的Hi-DER模型,在三个层级(异常分类L1、低级别或高级别分类L2、具体分级分类L3)评估关节病分类。模型在内部测试中加权平均AUC为L1 0.994、L2 0.980、L3 0.973;髋关节外部验证AUC为L1 0.978、L2 0.977、L3 0.971;膝关节外部验证AUC为L1 0.934、L2 0.929、L3 0.857。结论认为Hi-DER可能通过准确分类多关节关节病分级提高诊断效率,并可能有助于早期治疗。

要点

  • 模型在三个层级分类关节病,内部测试加权平均AUC L1 0.994、L2 0.980、L3 0.973。
  • 髋关节外部验证加权平均AUC L1 0.978、L2 0.977、L3 0.971。
  • 膝关节外部验证加权平均AUC L1 0.934、L2 0.929、L3 0.857。

核心结论

  • Hi-DER可能通过准确分类多关节关节病分级提高关节病诊断效率,并可能有助于早期治疗。

研究设计与样本

  • 本研究为方法学开发与外部验证研究,共纳入来自三家三级医院的1371张膝、肘、踝、肩、髋关节X线片。
  • 模型开发使用Sinchon Severance Hospital于2022年7月1日至12月31日收集的934张膝、肘、踝、肩关节X线片。
  • 外部验证使用Yongin Severance Hospital于2022年1月1日至12月31日收集的125张髋关节X线片,以及Gangnam Severance Hospital于2023年1月1日至6月31日收集的312例膝关节病例。
  • Hi-DER模型在四个关节上采用五折交叉验证进行逐步训练。
  • 关节病分类在三个层级评估:异常分类(L1)、低级别或高级别分类(L2)、具体分级分类(L3)。

诊断标准与关键数值

  • 内部测试加权平均AUC:L1 0.994(95% CI: 0.985, 0.999)、L2 0.980(95% CI: 0.958, 0.996)、L3 0.973(95% CI: 0.943, 0.993)。
  • 输入区域擦除70%时模型AUC仍保持在0.800以上。
  • 髋关节外部验证加权平均AUC:L1 0.978(95% CI: 0.952, 0.996)、L2 0.977(95% CI: 0.946, 0.996)、L3 0.971(95% CI: 0.934, 0.996)。
  • 膝关节外部验证加权平均AUC:L1 0.934(95% CI: 0.904, 0.958)、L2 0.929(95% CI: 0.900, 0.954)、L3 0.857(95% CI: 0.816, 0.894)。

临床应用与阅片要点

  • 模型性能与另外两个AI模型及三名放射科医生的分级预测进行了比较。
  • 模型可解释性采用梯度加权类激活映射(Grad-CAM)和渐进擦除加渐进恢复(PEPPR)方法。

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

PURPOSE: To develop a hierarchical continual arthropathy classification model for multiple joints that can be updated continuously for large-scale studies of various anatomical structures.

MATERIALS AND METHODS: This study included a total of 1371 radiographs of knee, elbow, ankle, shoulder, and hip joints from three tertiary hospitals. For model development, 934 radiographs of the knee, elbow, ankle, and shoulder were gathered from Sinchon Severance Hospital between July 1 and December 31, 2022. For external validation, 125 hip radiographs were collected from Yongin Severance Hospital between January 1 and December 31, 2022, and 312 knee cases were gathered from Gangnam Severance Hospital between January 1 and June 31, 2023. The Hierarchical Dynamically Expandable Representation (Hi-DER) model was trained stepwise on four joints using five-fold cross-validation. Arthropathy classification was evaluated at three hierarchical levels: abnormal classification (L1), low-grade or high-grade classification (L2), and specific grade classification (L3). The model's performance was compared with the grading predictions of two other AI models and three radiologists. For model explainability, gradient-weighted class activation mapping (Grad-CAM) and progressive erasing plus progressive restoration (PEPPR) were employed.

RESULTS: The model achieved a weighted average AUC of 0.994 (95% CI: 0.985, 0.999) for L1, 0.980 (95% CI: 0.958, 0.996) for L2, and 0.973 (95% CI: 0.943, 0.993) for L3. The model maintained an AUC above 0.800 with 70% of the input regions erased. During external validation on hip joints, the model demonstrated a weighted average AUC of 0.978 (95% CI: 0.952, 0.996) for L1, 0.977 (95% CI: 0.946, 0.996) for L2, and 0.971 (95% CI: 0.934, 0.996) for L3. For external knee data, the model yielded a weighted average AUC of 0.934 (95%: CI 0.904, 0.958), 0.929 (95% CI: 0.900, 0.954), and 0.857 (95% CI: 0.816, 0.894) for L1, L2, and L3, respectively.

CONCLUSION: The Hi-DER may enhance the efficiency of arthropathy diagnosis through accurate classification of arthropathy grades across multiple joints, potentially enabling early treatment.

来源:Radiol Med,查看原文。版权归原作者所有。

引用

  1. Jang BK, Kim S, Yu JY, et al. Classification models for arthropathy grades of multiple joints based on hierarchical continual learning. Radiol Med. 2025. PMID: 40126794. doi:10.1007/s11547-025-01974-4

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。