开放权重深度学习用于根尖周指数评分:三种卷积神经网络架构及集成方法的训练与比较
本研究从口内X线片提取17,549个根尖中心图像片段,由校准观察者标注PAI评分(1-5),训练ResNet50、EfficientNet-B3和ConvNeXt-Tiny三种CNN进行PAI评分。集成模型在200张独立平衡测试集上QWK最高为0.79,但置信区间重叠,未显著优于单个架构;二分类性能相当,集成模型准确率86%。所有模型以MIT许可证开源。
要点
- 集成模型QWK为0.79,EfficientNet-B3为0.77,ResNet50为0.76,ConvNeXt-Tiny为0.70,置信区间重叠。
- 二分类(健康PAI 1-2 vs 病变PAI 3-5)集成模型准确率86%、敏感度83%、特异度90%。
- 集成模型未显著优于单个CNN架构,三种架构性能相当。
核心结论
- 三种CNN模型及集成模型在五分类和二分类PAI评分上性能相当,集成模型未显著优于单个架构。
研究设计与样本
- 数据集为17,549个根尖中心图像片段,从口内X线片提取,由校准观察者标注PAI评分(1-5)。
- 图像片段按80/20分为训练集和验证集,基于验证集QWK选择模型。
- 训练ResNet50、EfficientNet-B3和ConvNeXt-Tiny三种CNN,并组合为等权重软投票集成模型。
- 所有模型在200张独立平衡测试集(每个PAI评分40张)上评估,QWK为主要指标,bootstrap估计置信区间。
主要发现
- 集成模型测试集QWK为0.79,EfficientNet-B3为0.77,ResNet50为0.76,ConvNeXt-Tiny为0.70,置信区间重叠。
- 二分类PAI分类中,三种CNN模型和集成模型性能相当。
- 集成模型二分类结果:准确率86%,敏感度83%,特异度90%,PPV 93%,NPV 78%,F1分数0.88。
关键数值与推荐
- PAI评分采用1-5分制,二分类将PAI 1-2定义为健康,PAI 3-5定义为病变。
- 主要结局指标为二次加权kappa(QWK),采用bootstrap估计置信区间。
临床怎么用
- AI模型可用于口内X线片的PAI评分辅助,集成模型与单个CNN性能相当,可根据资源选择单一架构。
- 模型以MIT许可证开源,便于临床研究和进一步验证。
局限性
- 测试集仅200张图像(每个PAI评分40张),样本量有限。
- 集成模型未显示统计学显著改善,置信区间重叠。
- 图像片段由单一校准观察者标注,未报告观察者间一致性。
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
AIM: This study aimed to (i) train and compare three convolutional neural network (CNN) architectures for Periapical Index (PAI) scoring on intraoral radiographs, and (ii) assess whether an ensemble of these models improves performance compared with any single architecture. METHODS: A dataset of 17,549 apex-centred image clips extracted from intraoral radiographs was annotated by a calibrated observer (KW) with PAI scores (1-5) and used to train three CNN architectures (ResNet50, EfficientNet-B3 and ConvNeXt-Tiny) for PAI scoring. The image clips were split 80/20 into training and validation sets, and model selection was based on validation performance of quadratic weighted kappa (QWK). The three CNN models were combined into an equal-weight soft-voting ensemble model, and all four models were evaluated on an independent balanced test set of 200 images (40 per PAI score). QWK was used as the primary outcome metric, and bootstrapping was used to estimate the confidence intervals (CIs). The PAI scale was dichotomized into healthy (PAI 1-2) and diseased (PAI 3-5), and accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and F1 score were calculated. RESULTS: The ensemble model achieved the highest performance with a QWK of 0.79 on the test set, followed by EfficientNet-B3 (QWK = 0.77), ResNet50 (QWK = 0.76) and ConvNeXt-Tiny (QWK = 0.70), with overlapping confidence intervals (CIs). For the binary PAI classification, the performance of the three CNN models and the ensemble model was comparable. The ensemble model achieved the following results: accuracy 86%, sensitivity 83%, specificity 90%, PPV 93%, NPV 78% and F1 score 0.88. CONCLUSION: Optimization and evaluation of three CNN models and an ensemble model for PAI scoring showed comparable performance on both the five-class PAI scale and the binary PAI scale on an independent test set. The ensemble model did not show statistically significant improvement over the three individual CNN architectures. The AI models trained and tested in this study are made open-source under the open-source MIT licence.
引用
- Torgersen G, Winsnes K, Handal T, et al. Open-Weight Deep Learning for Periapical Index Scoring: Training and Comparison of Convolutional Neural Network Models Based on Three Distinct Architectures and an Ensemble Approach. Int Endod J. 2026. PMID: 42751797. doi:10.1111/iej.70268
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。