文献速递口腔黏膜病口腔颌面外科

深度学习系统预测口腔白斑上皮异常增生

A Deep Learning System to Predict Epithelial Dysplasia in Oral Leukoplakia.
PubMed 文献 · J Dent Res · 2024-10-09

本研究基于口腔照片构建深度学习模型,用于预测口腔白斑(OL)患者的上皮异常增生(OED)概率,并与临床医生评分比较及评估决策支持能力。模型在测试集和外部验证中均表现出良好的区分度与校准度,且外部验证中模型分类OED状态的准确率优于92.3%的人类评分者。

要点

  • 基于口腔照片的深度学习模型可预测白斑上皮异常增生概率与状态。
  • 模型在外部验证中区分度良好,AUC为0.828(0.792-0.864)。
  • 模型分类OED状态的准确率优于92.3%的人类评分者。

核心结论

  • 基于照片的智能模型能以良好的校准度和区分度预测白斑的OED概率和状态,有望支持活检决策、避免不必要的活检并辅助患者自我监测。

研究设计与样本

  • 回顾性研究,收集活检/组织病理学前的白斑照片构建深度学习模型(n=2,073),以组织病理学OED状态为金标准。
  • 开发并内部验证以EfficientNet-B2为骨干的深度学习架构,输出OED预测概率、OED状态和感兴趣区域热图。
  • 在时间队列上测试性能,随后进行地理外部验证,并与人类评分者对OED状态的判断进行比较。

主要发现

  • 测试集中模型Brier分数为0.124(0.079-0.169),AUC为0.882(0.838-0.926),平衡准确率为81.8%(76.5-87.1)。
  • 外部验证中模型Brier分数为0.146(0.112-0.18),AUC为0.828(0.792-0.864),平衡准确率为76.4%(72.3-80.5)。
  • 在OED评估中,模型选择白斑患者进行活检/组织病理学的潜在净获益高于对所有患者进行活检。
  • 外部验证显示,模型从口腔图像分类白斑OED状态的准确率优于92.3%(24/26)的人类评分者(人类评分者平衡准确率:54.8%-79.7%)。

临床怎么用

  • 该模型可作为决策支持工具,帮助临床医生选择需要活检/组织病理学检查的白斑患者,减少不必要的活检。
  • 模型预测结果可辅助患者自我监测,但需结合临床评估。

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

Oral leukoplakia (OL) has an inherent disposition to develop oral cancer. OL with epithelial dysplasia (OED) is significantly likely to undergo malignant transformation; however, routine OED assessment is invasive and challenging. This study investigated whether a deep learning (DL) model can predict dysplasia probability among patients with leukoplakia using oral photographs. In addition, we assessed the performance of the DL model in comparison with clinicians' ratings and in providing decision support on dysplasia assessment. Retrospective images of leukoplakia taken before biopsy/histopathology were obtained to construct the DL model (n = 2,073). OED status following histopathology was used as the gold standard for all images. We first developed, fine-tuned, and internally validated a DL architecture with an EfficientNet-B2 backbone that outputs the predicted probability of OED, OED status, and regions-of-interest heat maps. Then, we tested the performance of the DL model on a temporal cohort before geographical validation. We also assessed the model's performance at external validation with opinions provided by human raters on OED status. Performance evaluation included discrimination, calibration, and potential net benefit. The DL model achieved good Brier scores, areas under the curve, and balanced accuracies of 0.124 (0.079-0.169), 0.882 (0.838-0.926), and 81.8% (76.5-87.1) at testing and 0.146 (0.112-0.18), 0.828 (0.792-0.864), and 76.4% (72.3-80.5) at external validation, respectively. In addition, the model had a higher potential net benefit in selecting patients with OL for biopsy/histopathology during OED assessment than when biopsies were performed for all patients. External validation also showed that the DL model had better accuracy than 92.3% (24/26) of human raters in classifying the OED status of leukoplakia from oral images (balanced accuracy: 54.8%-79.7%). Overall, the photograph-based intelligent model can predict OED probability and status in leukoplakia with good calibration and discrimination, which shows potential for decision support to select patients for biopsy/histopathology, obviate unnecessary biopsy, and assist in patient self-monitoring.

来源:J Dent Res,查看原文。版权归原作者所有。

引用

  1. Adeoye J, Chaurasia A, Akinshipo A, et al. A Deep Learning System to Predict Epithelial Dysplasia in Oral Leukoplakia. J Dent Res. 2024. PMID: 39382109. doi:10.1177/00220345241272048

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。