Dehua Tang1, Muhan Ni1, Chang Zheng1, Xiwei Ding1, Nina Zhang1, Tian Yang1, Qiang Zhan2, Yiwei Fu3, Wenjia Liu4, Duanming Zhuang5, Ying Lv1, Guifang Xu6, Lei Wang7, Xiaoping Zou8. 1. Department of Gastroenterology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, 210008, Jiangsu, China. 2. Department of Gastroenterology, Wuxi People's Hospital, Affiliated Wuxi People's Hospital With Nanjing Medical University, Wuxi, 214023, Jiangsu, China. 3. Department of Gastroenterology, Taizhou People's Hospital, The Fifth Affiliated Hospital With Nantong University, Taizhou, 225300, Jiangsu, China. 4. Department of Gastroenterology, Changzhou Second People's Hospital, Changzhou, 213003, Jiangsu, China. 5. Department of Gastroenterology, Nanjing Gaochun People's Hospital, Nanjing, 211300, Jiangsu, China. 6. Department of Gastroenterology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, 210008, Jiangsu, China. 13852293376@163.com. 7. Department of Gastroenterology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, 210008, Jiangsu, China. 867152094@qq.com. 8. Department of Gastroenterology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, 210008, Jiangsu, China. zouxp@nju.edu.cn.
Abstract
BACKGROUND: Diagnosis of early gastric cancer (EGC) under narrow band imaging endoscopy (NBI) is dependent on expertise and skills. We aimed to elucidate whether artificial intelligence (AI) could diagnose EGC under NBI and evaluate the diagnostic assistance of the AI system. METHODS: In this retrospective diagnostic study, 21,785 NBI images and 20 videos from five centers were divided into a training dataset (13,151 images, 810 patients), an internal validation dataset (7057 images, 283 patients), four external validation datasets (1577 images, 147 patients), and a video validation dataset (20 videos, 20 patients). All the images were labeled manually and used to train an AI system using You look only once v3 (YOLOv3). Next, the diagnostic performance of the AI system and endoscopists were compared and the diagnostic assistance of the AI system was assessed. The accuracy, sensitivity, specificity, and AUC were primary outcomes. RESULTS: The AI system diagnosed EGCs on validation datasets with AUCs of 0.888-0.951 and diagnosed all the EGCs (100.0%) in video dataset. The AI system achieved better diagnostic performance (accuracy, 93.2%, 95% CI, 90.0-94.9%) than senior (85.9%, 95% CI, 84.2-87.4%) and junior (79.5%, 95% CI, 77.8-81.0%) endoscopists. The AI system significantly enhanced the performance of endoscopists in senior (89.4%, 95% CI, 87.9-90.7%) and junior (84.9%, 95% CI, 83.4-86.3%) endoscopists. CONCLUSION: The NBI AI system outperformed the endoscopists and exerted potential assistant impact in EGC identification. Prospective validations are needed to evaluate the clinical reinforce of the system in real clinical practice.
BACKGROUND: Diagnosis of early gastric cancer (EGC) under narrow band imaging endoscopy (NBI) is dependent on expertise and skills. We aimed to elucidate whether artificial intelligence (AI) could diagnose EGC under NBI and evaluate the diagnostic assistance of the AI system. METHODS: In this retrospective diagnostic study, 21,785 NBI images and 20 videos from five centers were divided into a training dataset (13,151 images, 810 patients), an internal validation dataset (7057 images, 283 patients), four external validation datasets (1577 images, 147 patients), and a video validation dataset (20 videos, 20 patients). All the images were labeled manually and used to train an AI system using You look only once v3 (YOLOv3). Next, the diagnostic performance of the AI system and endoscopists were compared and the diagnostic assistance of the AI system was assessed. The accuracy, sensitivity, specificity, and AUC were primary outcomes. RESULTS: The AI system diagnosed EGCs on validation datasets with AUCs of 0.888-0.951 and diagnosed all the EGCs (100.0%) in video dataset. The AI system achieved better diagnostic performance (accuracy, 93.2%, 95% CI, 90.0-94.9%) than senior (85.9%, 95% CI, 84.2-87.4%) and junior (79.5%, 95% CI, 77.8-81.0%) endoscopists. The AI system significantly enhanced the performance of endoscopists in senior (89.4%, 95% CI, 87.9-90.7%) and junior (84.9%, 95% CI, 83.4-86.3%) endoscopists. CONCLUSION: The NBI AI system outperformed the endoscopists and exerted potential assistant impact in EGC identification. Prospective validations are needed to evaluate the clinical reinforce of the system in real clinical practice.
Authors: Wenya Linda Bi; Ahmed Hosny; Matthew B Schabath; Maryellen L Giger; Nicolai J Birkbak; Alireza Mehrtash; Tavis Allison; Omar Arnaout; Christopher Abbosh; Ian F Dunn; Raymond H Mak; Rulla M Tamimi; Clare M Tempany; Charles Swanton; Udo Hoffmann; Lawrence H Schwartz; Robert J Gillies; Raymond Y Huang; Hugo J W L Aerts Journal: CA Cancer J Clin Date: 2019-02-05 Impact factor: 508.702