Evaluation of cleanliness in capsule endoscopy is essential to validate the exam. However, yet reliable evaluation of bowel cleanliness remains challenging. The validated KODA score provides robust assessment but is time-consuming and impractical in routine practice. We aimed to evaluate the performance of an AI-based tool (AXAROlite®, Augmented Endoscopy) compared with the KODA score for assessing small bowel cleanliness in Crohn's disease (CD) patients.

This  was a post-hoc analysis of a multicenter randomized controlled trial (NCT05117996) including 142 CD patients undergoing small bowel capsule endoscopy (SBCE) after either a standard PEG-based or simplified clear liquid preparation. All videos were evaluated for cleanliness using the KODA score by trained readers. The same videos were analyzed using the AI-based AXAROlite® tool. Correlation, agreement, and diagnostic performance of AXAROlite® were compared with the KODA score. Clinical factors influencing AI-assessed cleanliness were also investigated.

Correlation between the KODA score and the AI-AXAROlite® score was found strong for the whole SB (ρ=0.61; r2=0.35; p<0.001) and moderate to strong for each of the four quartiles analyzed: ρ=0.52; r2=0.24; p<0.001, ρ=0.56; r2=0.24; p<0.001, ρ=0.50; r2=0.21; p<0.001, ρ=0.62; r2=0.41; p<0.001 for the first, the second, the third and the fourth quartiles, respectively.When compared to the qualitative evaluation of the SB cleanliness i.e. good, fair or poor, both the KODA and the AXAROlite® scores decreased similarly. The median [IQR] values of AXAROlite® scores were respectively 75.0 [59.0-84.0], 56.0 [42.0-67.0] and 41.00 [25.75-59.75]. The differences were statistically significant between a good and a fair or poor cleanliness (p<0.001)AXAROlite® demonstrated excellent diagnostic accuracy for detecting adequate cleanliness defined as KODA > 2.25 (AUROC=0.85, 95% CI 0.79–0.92), with an optimal threshold of 72% clean frames (sensitivity 86%, specificity 72%).In multivariate analysis, the activity of the disease and the SB transit time influenced negatively the AXAROlite® score: β=-7.87 [-15.29; -0.44], p= 0.0382 and β=-0.04 [-0.06; -0.01], p= 0.0031 respectively. The behavior, the location of the disease, the endoscopic severity and the presence of an anastomosis, did not modify independently the AXAROlite® SBscore.

AXAROlite® provides a rapid, fully automated, and accurate evaluation of small bowel cleanliness in CD, comparable to the validated KODA score. Its integration into routine workflows may streamline reporting and reduce inter-observer variability.