文献速递咽喉与嗓音头颈外科

基于白光与窄带成像的多实例学习用于声带白斑诊断:一项多中心研究

Multi-Instance Learning for Vocal Fold Leukoplakia Diagnosis Using White Light and Narrow-Band Imaging: A Multicenter Study.
PubMed 文献 · Laryngoscope · 2024-05-27

该多中心研究利用426例患者的7057张白光和窄带成像图像开发并内部验证了多实例学习AI模型,另用两家医院的1617张图像进行外部验证。模型内部和外部验证AUC分别达0.868和0.884,视频验证AUC为0.825,人机对比显示AI显著提升所有喉科医生的AUC和准确率。

要点

  • 多实例学习AI模型融合白光和窄带成像,内部验证AUC 0.868、外部验证AUC 0.884。
  • 视频验证集AUC为0.825(95%CI 0.704-0.946)。
  • 人机对比中AI显著提高所有喉科医生的AUC和准确率(p<0.05)。

核心结论

  • 多中心研究开发的基于多实例学习和白光与窄带成像融合的AI模型对声带白斑诊断有效,尤其有助于低年资喉科医生。
  • 但需进一步优化和验证以全面评估其在临床环境中的潜在影响。

研究设计与样本

  • 多中心研究,使用426例患者的7057张图像进行模型开发和内部验证,另用两家医院的1617张图像进行外部验证。
  • 采用深度学习结合多实例学习方法基于白光和窄带成像模态建模,并用50段前瞻性收集的视频评估实时性能。
  • 通过涉及100例患者和12名喉科医生的人机对比评估模型真实世界效果。

主要发现

  • 模型在内部和外部验证集的最高AUC分别为0.868和0.884。
  • 视频验证集AUC为0.825(95%CI:0.704-0.946)。
  • 人机对比中,AI显著提高所有喉科医生的AUC和准确率(p<0.05)。
  • 在AI辅助下,所有喉科医生的诊断能力和一致性均提高。

关键数值与推荐

  • 内部验证集AUC 0.868,外部验证集AUC 0.884。
  • 视频验证集AUC 0.825(95%CI:0.704-0.946)。
  • 人机对比AI显著提高所有喉科医生AUC和准确率(p<0.05)。

临床怎么用

  • 作者认为该AI模型对声带白斑诊断有效,尤其有助于低年资喉科医生。
  • 作者指出需进一步优化和验证以全面评估其临床潜在影响。

局限性

  • 需进一步优化和验证以全面评估其临床环境中的潜在影响。

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: Vocal fold leukoplakia (VFL) is a precancerous lesion of laryngeal cancer, and its endoscopic diagnosis poses challenges. We aim to develop an artificial intelligence (AI) model using white light imaging (WLI) and narrow-band imaging (NBI) to distinguish benign from malignant VFL. METHODS: A total of 7057 images from 426 patients were used for model development and internal validation. Additionally, 1617 images from two other hospitals were used for model external validation. Modeling learning based on WLI and NBI modalities was conducted using deep learning combined with a multi-instance learning approach (MIL). Furthermore, 50 prospectively collected videos were used to evaluate real-time model performance. A human-machine comparison involving 100 patients and 12 laryngologists assessed the real-world effectiveness of the model. RESULTS: The model achieved the highest area under the receiver operating characteristic curve (AUC) values of 0.868 and 0.884 in the internal and external validation sets, respectively. AUC in the video validation set was 0.825 (95% CI: 0.704-0.946). In the human-machine comparison, AI significantly improved AUC and accuracy for all laryngologists (p < 0.05). With the assistance of AI, the diagnostic abilities and consistency of all laryngologists improved. CONCLUSIONS: Our multicenter study developed an effective AI model using MIL and fusion of WLI and NBI images for VFL diagnosis, particularly aiding junior laryngologists. However, further optimization and validation are necessary to fully assess its potential impact in clinical settings. LEVEL OF EVIDENCE: 3 Laryngoscope, 134:4321-4328, 2024.

来源:Laryngoscope,查看原文。版权归原作者所有。

引用

  1. Tie CW, Li DY, Zhu JQ, et al. Multi-Instance Learning for Vocal Fold Leukoplakia Diagnosis Using White Light and Narrow-Band Imaging: A Multicenter Study. Laryngoscope. 2024. PMID: 38801129. doi:10.1002/lary.31537

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。