Medical AI company Ainex Corporation (CEO David Lee) has released findings from a multi-center study using its AI-based real-time colonoscopy diagnostic assistance software, ENAD CADx.

According to the company on the 29th, the study examined the impact of AI diagnostic assistance software on endoscopists' diagnostic accuracy and has been published in the international gastroenterology journal Gut and Liver.

Results confirmed that ENAD CADx assistance significantly improved endoscopist performance, with the greatest effect observed in the novice group. This is meaningful as it demonstrates AI's potential to reduce variation in optical diagnosis quality attributable to differences in endoscopist skill level.

Optical diagnosis — the real-time observation of lesion morphology during endoscopy to estimate histological type — has drawn attention for its ability to reduce the risks and costs of unnecessary polyp resection.

However, clinical application requires meeting international benchmarks including SODA (Simple Optical Diagnosis Accuracy) and PIVI (Preservation and Incorporation of Valuable Endoscopic Innovations). Given the high diagnostic accuracy and negative predictive value (NPV) these standards demand, they are considered difficult targets for novice endoscopists to reach.

The study enrolled 35 endoscopists from four institutions: Seoul National University Hospital Gangnam Center, Seoul National University Hospital, Pusan National University Hospital, and Inje University Busan Paik Hospital. Participants were classified into three groups — trained experts, experts, and novices — based on their optical diagnosis training and experience. Each evaluated 100 colorectal lesions (51 adenomas, 39 hyperplastic polyps, 10 sessile serrated lesions) presented as narrow-band imaging (NBI) video clips, with and without ENAD CADx assistance.

Findings showed that ENAD CADx assistance enabled faster, higher-confidence judgments based on real-time AI analysis, leading to improved diagnostic accuracy.

The research team defined a "high-confidence (HC) diagnosis" as a judgment made without hesitation within 3 seconds of observing a lesion. With ENAD CADx, HC diagnostic accuracy rose from 78.3% to 89.8% — an 11.5-percentage-point improvement. Average decision-making time also decreased from 3.3 to 2.8 seconds, reflecting enhanced diagnostic efficiency.

The effect was most pronounced in the novice group, where HC diagnostic accuracy improved from 74.1% to 88.8% — a gain of approximately 14.7 percentage points — and inter-rater diagnostic agreement increased significantly as well. This suggests ENAD CADx can serve as a "training equalizer," reducing experience-driven variation in diagnostic quality.

ENAD CADx was also found to simultaneously improve both the accuracy and rate of high-confidence diagnoses, suggesting it could help endoscopists meet PIVI and SODA benchmarks — further enhancing the feasibility of optical diagnosis in real-world clinical settings.

David Lee, CEO of Ainex, stated: "This multi-center study provides important evidence that ENAD CADx goes beyond simply improving diagnostic accuracy — it can contribute to raising the overall standard of endoscopy quality. Through continued clinical research, we will establish ENAD CADx as a trusted endoscopy AI solution in the global market and set a new standard for colonoscopy diagnosis."