The colonoscopist-blinded, randomized study included 1,434 patients, according to Becker’s. Participants were broken into two groups, the first group, labeled as the control group, performed routine self-evaluation throughout their preparation process, and a group in which the tool was used to gauge effectiveness. The second group used the AI-based tool to assist their preparation.
“The primary outcome was the consistency (homogeneity) between the results of the two methods,” the study states. The secondary outcomes included the quality of bowel preparation according to the Boston Bowel Preparation Scale (BBPS), the polyp detection rate (PDR), and the adenoma detection rate (ADR).”
Results of the study showed that pass/no pass rates between the groups were similar, as well as polyp detection rate (PDR), adenoma detection rate (ADR), and the quality of the prep based on the Boston Bowel Preparation Scale (BBPS). The main differentiator in the results was the ranking of mean BBPS scores for those who passed.
“The mean BBPS score of patients with “pass” results were signiﬁcantly higher in the AI-CNN group than in the Control group, indicating that the AI-CNN model may further improve the quality of bowel preparation in patients exhibiting adequate bowel preparation,” the study said.
Results showed that while the tool provided roughly the same percentage of passing results, the quality of those results were significantly improved, providing higher quality preparation for patients.
To read the study and its results, click here.