Artificial intelligence versus expert endoscopists for diagnosis of gastric cancer in patients who have undergone upper gastrointestinal endoscopy

Abstract

Introduction and aims: To compare endoscopy gastric cancer images diagnosis rate between artificial intelligence (AI) and expert endoscopists. Patients and methods: We used the retrospective data of 500 patients, including 100 with gastric cancer, matched 1:1 to diagnosis by AI or expert endoscopists. We retrospectively evaluated the non-inferiority (prespecified margin 5%) of the per-patient rate of gastric cancer diagnosis by AI and compared the per-image rate of gastric cancer diagnosis. Results: Gastric cancer was diagnosed in 49 of 49 patients (100%) in the AI group and 48 of 51 patients (94.12%) in the expert endoscopist group (difference 5.88, 95% confidence interval: −0.58 to 12.3). The per-image rate of gastric cancer diagnosis was higher in the AI group (99.87%, 747/748 images) than in the expert endoscopist group (88.17%, 693/786 images) (difference 11.7%). Conclusions: Non inferiority of the rate of gastric cancer diagnosis by AI was demonstrated but superiority is not demonstrated.

Response

“…Yang et al 1 referred to "visual pathology without tissue biopsy" with the use of artificial intelligence (AI) models based on endoscopy and "liquid biopsy." Indeed, recent studies 3,4 have established AI-based endoscopic detection systems and also cell-free DNA-based detection systems for GC. However, we would gently emphasize that detection and prediction are different concepts, and that what is essential for the prediction of GC is a system that can detect and diagnose gastric precancerous conditions (ie, gastric atrophy and intestinal metaplasia), as recently reported.…”

Endoscopic Classifications of Early Gastric Cancer: A Literature Review

“…One study by Hirasawa et al, which utilized convolutional neural networks to detect EGC in endoscopic images, showed that the sensitivity of AI for white light endoscopy and magnifying NBI were 92% and 97%, respectively [ 35 ]. Meanwhile, a study done by Niikura et al did not show inferiority nor superiority of AI when compared to expert endoscopists [ 36 ]. However, a meta-analysis study showed that the sensitivity and specificity of AI in detecting EGC was 86% and 93%, respectively, concluding that AI was more accurate in detecting EGC compared to expert endoscopists [ 37 ].…”

Artificial intelligence in endoscopy: transition from academic development to integration into daily endoscopic practice

“…Lack of generalizability of AI performance is a widespread problem. Virtually all endoscopic AI studies, including the paper by Niikura and colleagues 4 , rely on heavily curated, high-quality imagery collected at high-volume academic centers by expert endoscopists. However, the quality of imaging procedures in community hospitals, where the bulk of surveillance endoscopies are performed, and where therefore – paradoxically – most AI systems will be employed, is considerably lower.…”

“…In this issue of Endoscopy, Niikura and colleagues present results of a retrospective study in which they evaluated a previously trained computer-aided detection system for early gastric cancer [4]. For this evaluation, they applied an interesting experimental set-up using stratified matching of retrospectively collected imagery in a noninferiority comparison of AI versus expert assessment.…”

“…Given its subtle endoscopic appearance and the corresponding diagnostic challenge, early gastric cancer is a logical application of AI. Several systems have recently been described, mainly focusing on primary detection systems [1][2][3].In this issue of Endoscopy, Niikura and colleagues present results of a retrospective study in which they evaluated a previously trained computer-aided detection system for early gastric cancer [4]. For this evaluation, they applied an interesting experimental set-up using stratified matching of retrospectively collected imagery in a noninferiority comparison of AI versus expert assessment.…”