Department of Endoscopy, National Cancer Center, National Clinical Research Center for Cancer, Cancer Hospital & Shenzhen Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Shenzhen, China.
Department of Endoscopy, National Cancer Center, National Clinical Research Center for Cancer, Cancer Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, China.
Am J Otolaryngol. 2024 Jul-Aug;45(4):104342. doi: 10.1016/j.amjoto.2024.104342. Epub 2024 Apr 30.
To develop a multi-instance learning (MIL) based artificial intelligence (AI)-assisted diagnosis models by using laryngoscopic images to differentiate benign and malignant vocal fold leukoplakia (VFL).
The AI system was developed, trained and validated on 5362 images of 551 patients from three hospitals. Automated regions of interest (ROI) segmentation algorithm was utilized to construct image-level features. MIL was used to fusion image level results to patient level features, then the extracted features were modeled by seven machine learning algorithms. Finally, we evaluated the image level and patient level results. Additionally, 50 videos of VFL were prospectively gathered to assess the system's real-time diagnostic capabilities. A human-machine comparison database was also constructed to compare the diagnostic performance of otolaryngologists with and without AI assistance.
In internal and external validation sets, the maximum area under the curve (AUC) for image level segmentation models was 0.775 (95 % CI 0.740-0.811) and 0.720 (95 % CI 0.684-0.756), respectively. Utilizing a MIL-based fusion strategy, the AUC at the patient level increased to 0.869 (95 % CI 0.798-0.940) and 0.851 (95 % CI 0.756-0.945). For real-time video diagnosis, the maximum AUC at the patient level reached 0.850 (95 % CI, 0.743-0.957). With AI assistance, the AUC improved from 0.720 (95 % CI 0.682-0.755) to 0.808 (95 % CI 0.775-0.839) for senior otolaryngologists and from 0.647 (95 % CI 0.608-0.686) to 0.807 (95 % CI 0.773-0.837) for junior otolaryngologists.
The MIL based AI-assisted diagnosis system can significantly improve the diagnostic performance of otolaryngologists for VFL and help to make proper clinical decisions.
利用喉镜图像开发基于多实例学习（MIL）的人工智能（AI）辅助诊断模型，以区分良性和恶性声带白斑（VFL）。
该 AI 系统是在来自三家医院的 551 名患者的 5362 张图像上开发、训练和验证的。利用自动感兴趣区域（ROI）分割算法构建图像级特征。使用 MIL 融合图像级结果以获得患者级特征，然后使用七种机器学习算法对提取的特征进行建模。最后，我们评估了图像级和患者级结果。此外，前瞻性收集了 50 个 VFL 视频，以评估系统的实时诊断能力。还构建了人机比较数据库，以比较 AI 辅助和不辅助的耳鼻喉科医生的诊断性能。
在内部和外部验证集中，图像级分割模型的最大曲线下面积（AUC）分别为 0.775（95%CI 0.740-0.811）和 0.720（95%CI 0.684-0.756）。利用基于 MIL 的融合策略，患者级 AUC 提高至 0.869（95%CI 0.798-0.940）和 0.851（95%CI 0.756-0.945）。对于实时视频诊断，患者级 AUC 的最大值达到 0.850（95%CI，0.743-0.957）。在 AI 辅助下，高级耳鼻喉科医生的 AUC 从 0.720（95%CI 0.682-0.755）提高到 0.808（95%CI 0.775-0.839），初级耳鼻喉科医生的 AUC 从 0.647（95%CI 0.608-0.686）提高到 0.807（95%CI 0.773-0.837）。
基于 MIL 的 AI 辅助诊断系统可以显著提高耳鼻喉科医生对 VFL 的诊断性能，并有助于做出适当的临床决策。
Zotero 插件