Real-time use of artificial intelligence for diagnosing early gastric cancer by magnifying image-enhanced endoscopy: a multicenter diagnostic study (with videos)

Post written by Honggang Yu, MD, from the Department of Gastroenterology, Key Laboratory of Hubei Province for Digestive System Disease, and Hubei Provincial Clinical Research Center for Digestive Disease Minimally Invasive Incision, Renmin Hospital of Wuhan University, Wuhan, China.

Yu_photo

In this study, we used deep learning models to develop an artificial intelligence system called ENDOANGEL-ME to diagnose early gastric cancer (EGC) in magnifying image-enhanced endoscopy. ENDOANGEL-ME achieved good and stable diagnostic performance at 3 levels, including retrospective static images, real-time videos, and a prospective clinical trial.

Early detection, diagnosis, and treatment are key strategies for improving survival of a gastric cancer patient. Endoscopy is the most common and important method for detecting EGC. However, EGCs are difficult to be recognized because of the subtle mucosa changes. The paucity of experts makes endoscopic recognition of EGC a great challenge in the clinic.

The accuracy of ENDOANGEL-ME for diagnosing EGC was 88.44% and 90.49% in internal and external images, respectively. In 93 internal videos, ENDOANGEL-ME achieved an accuracy of 90.32% for diagnosing EGC, significantly superior to that of senior endoscopists (70.16% ± 8.78).

In 94 external videos, with the assistance of ENDOANGEL-ME, endoscopists showed improved accuracy and sensitivity (85.64% vs 80.32% and 82.03% vs 67.19%). In 194 prospective consecutive patients with 251 lesions, ENDOANGEL-ME achieved a sensitivity of 92.59% (25/27) and an accuracy of 83.67% (210/251) in real clinical practice.

The results of the multicenter diagnostic study show that ENDOANGEL-ME could be well applied in the clinical setting. The clinical trial was conducted in 1 hospital in the present study. Real-time use of ENDOANGEL-ME in multicenters should be further investigated to provide sufficient evidence for its performance in clinical practice. 

In conclusion, computer-aided diagnosis of EGC is a promising research field in the foreseeable future.

Yu_figure

Workflow chart for the training, validation and test of ENDOANGEL-ME. M-IEE, Magnifying image-enhanced endoscopy; EGC, early gastric cancer; RHWU, Renmin Hospital of Wuhan University; CHW, Central Hospital of Wuhan; PHCTG, the People’s Hospital of China Three Gorges University; YCPH, Yichang Central People’s Hospital; JPH, Jingmen Petrochemical Hospital; XCH, Xiaogan Central Hospital; PUCH, Peking University Cancer Hospital.

Read the full article online.

The information presented in Endoscopedia reflects the opinions of the authors and does not represent the position of the American Society for Gastrointestinal Endoscopy (ASGE). ASGE expressly disclaims any warranties or guarantees, expressed or implied, and is not liable for damages of any kind in connection with the material, information, or procedures set forth.

Leave a Comment

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s