Yuan Peng, Liu Meichen, He Hangzhou, Dai Liang, Wu Ya-Ya, Chen Ke-Neng, Wu Qi, Lu Yanye
State Key Laboratory of Holistic Integrative Management of Gastrointestinal Cancers, Beijing Key Laboratory of Carcinogenesis and Translational Research, Endoscopy Center, Peking University Cancer Hospital and Institute, Beijing, China.
Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education), Endoscopy Center, Peking University Cancer Hospital and Institute, Peking University School of Oncology, Beijing, China.
Front Oncol. 2025 May 6;15:1590448. doi: 10.3389/fonc.2025.1590448. eCollection 2025.
BACKGROUND AND AIMS: Esophageal cancer (EC) patients may achieve pathological complete response (pCR) after receiving total neoadjuvant therapy (TNT), which allows them to avoid surgery and preserve organs. We aimed to benchmark the performance of existing artificial intelligence (AI) methods and develop a more accurate model for evaluating EC patients' response after TNT. METHODS: We built the Beijing-EC-TNT dataset, consisting of 7,359 images from 300 EC patients who underwent TNT at Beijing Cancer Hospital. The dataset was divided into Cohort1 (4,561 images, 209 patients) for cross-validation and Cohort 2 (2,798 images, 91 patients) for external evaluation. Patients and endoscopic images were labeled as either pCR or non-pCR based on postoperative pathology results. We systematically evaluated mainstream AI models and proposed EC-HAENet, a hybrid-architecture ensembled deep learning model. RESULTS: In image-level classification, EC-HAENet achieved an area under the curve of 0.98 in Cohort 1 and 0.99 in Cohort 2. In patient-level classification, the accuracy of EC-HAENet was significantly higher than that of endoscopic biopsy in both Cohorts 1 and 2 (accuracy, 0.93 . 0.78, P<0.0001 and 0.93 . 0.71, P<0.0001). CONCLUSION: EC-HAENet can assist endoscopists in accurately evaluating the response of EC patients after TNT.
背景与目的：食管癌（EC）患者在接受全新辅助治疗（TNT）后可能实现病理完全缓解（pCR），这使他们能够避免手术并保留器官。我们旨在评估现有人工智能（AI）方法的性能，并开发一个更准确的模型来评估EC患者在TNT后的反应。 方法：我们构建了北京-EC-TNT数据集，该数据集由在北京肿瘤医院接受TNT的300例EC患者的7359张图像组成。该数据集被分为用于交叉验证的队列1（4561张图像，209例患者）和用于外部评估的队列2（2798张图像，91例患者）。根据术后病理结果，将患者和内镜图像标记为pCR或非pCR。我们系统地评估了主流AI模型，并提出了EC-HAENet，一种混合架构集成深度学习模型。 结果：在图像水平分类中，EC-HAENet在队列1中的曲线下面积为0.98，在队列2中为0.99。在患者水平分类中，EC-HAENet在队列1和队列2中的准确率均显著高于内镜活检（准确率分别为0.93对0.78，P<0.0001；0.93对0.71，P<0.0001）。 结论：EC-HAENet可以协助内镜医师准确评估EC患者在TNT后的反应。
Clin Transl Gastroenterol. 2021-8-4
Zotero 插件
