Explainable artificial intelligence incorporated with domain knowledge diagnosing early gastric neoplasms under white light endoscopy

Zehua Dong; Junxiao Wang; Yanxia Li; Yunchao Deng; Wei Zhou; Xiaoquan Zeng; Dexin Gong; Jun Liu; Jie Pan; Renduo Shang; Youming Xu; Ming Xu; Lihui Zhang; Mengjiao Zhang; Xiao Tao; Yijie Zhu; Hongliu Du; Zihua Lu; Liwen Yao; Lianlian Wu; Honggang Yu

doi:10.1038/s41746-023-00813-y

Explainable artificial intelligence incorporated with domain knowledge diagnosing early gastric neoplasms under white light endoscopy

NPJ Digit Med. 2023 Apr 12;6(1):64. doi: 10.1038/s41746-023-00813-y.

Authors

Zehua Dong^#^{1

2

3}, Junxiao Wang^#^{1

2

3}, Yanxia Li^#^{1

2

3}, Yunchao Deng^{1

2

3}, Wei Zhou^{1

2

3}, Xiaoquan Zeng^{1

2

3}, Dexin Gong^{1

2

3}, Jun Liu^{1

2

3}, Jie Pan⁴, Renduo Shang^{1

2

3}, Youming Xu^{1

2

3}, Ming Xu^{1

2

3}, Lihui Zhang^{1

2

3}, Mengjiao Zhang⁵, Xiao Tao^{1

2

3}, Yijie Zhu^{1

2

3}, Hongliu Du^{1

2

3}, Zihua Lu^{1

2

3}, Liwen Yao^{1

2

3}, Lianlian Wu^{6

7

8}, Honggang Yu^{9

10

11}

Affiliations

¹ Renmin Hospital of Wuhan University, Wuhan, China.
² Key Laboratory of Hubei Province for Digestive System Disease, Renmin Hospital of Wuhan University, Wuhan, China.
³ Hubei Provincial Clinical Research Center for Digestive Disease Minimally Invasive Incision, Renmin Hospital of Wuhan University, Wuhan, China.
⁴ Department of Gastroenterology, Wenzhou Central Hospital, Wenzhou, China.
⁵ Department of Gastroenterology, The Central Hospital of Wuhan, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
⁶ Renmin Hospital of Wuhan University, Wuhan, China. wu_leanne@163.com.
⁷ Key Laboratory of Hubei Province for Digestive System Disease, Renmin Hospital of Wuhan University, Wuhan, China. wu_leanne@163.com.
⁸ Hubei Provincial Clinical Research Center for Digestive Disease Minimally Invasive Incision, Renmin Hospital of Wuhan University, Wuhan, China. wu_leanne@163.com.
⁹ Renmin Hospital of Wuhan University, Wuhan, China. yuhonggang@whu.edu.cn.
¹⁰ Key Laboratory of Hubei Province for Digestive System Disease, Renmin Hospital of Wuhan University, Wuhan, China. yuhonggang@whu.edu.cn.
¹¹ Hubei Provincial Clinical Research Center for Digestive Disease Minimally Invasive Incision, Renmin Hospital of Wuhan University, Wuhan, China. yuhonggang@whu.edu.cn.

^# Contributed equally.

Abstract

White light endoscopy is the most pivotal tool for detecting early gastric neoplasms. Previous artificial intelligence (AI) systems were primarily unexplainable, affecting their clinical credibility and acceptability. We aimed to develop an explainable AI named ENDOANGEL-ED (explainable diagnosis) to solve this problem. A total of 4482 images and 296 videos with focal lesions from 3279 patients from eight hospitals were used for training, validating, and testing ENDOANGEL-ED. A traditional sole deep learning (DL) model was trained using the same dataset. The performance of ENDOANGEL-ED and sole DL was evaluated on six levels: internal and external images, internal and external videos, consecutive videos, and man-machine comparison with 77 endoscopists in videos. Furthermore, a multi-reader, multi-case study was conducted to evaluate the ENDOANGEL-ED's effectiveness. A scale was used to compare the overall acceptance of endoscopists to traditional and explainable AI systems. The ENDOANGEL-ED showed high performance in the image and video tests. In man-machine comparison, the accuracy of ENDOANGEL-ED was significantly higher than that of all endoscopists in internal (81.10% vs. 70.61%, p < 0.001) and external videos (88.24% vs. 78.49%, p < 0.001). With ENDOANGEL-ED's assistance, the accuracy of endoscopists significantly improved (70.61% vs. 79.63%, p < 0.001). Compared with the traditional AI, the explainable AI increased the endoscopists' trust and acceptance (4.42 vs. 3.74, p < 0.001; 4.52 vs. 4.00, p < 0.001). In conclusion, we developed a real-time explainable AI that showed high performance, higher clinical credibility, and acceptance than traditional DL models and greatly improved the diagnostic ability of endoscopists.