Artificial intelligence systems could help lower time-related degradation of colonoscopy quality, according to a recent study.
The study, published Jan. 31 in JAMA Network Open, examined 1,780 colonoscopy procedures and found that adenoma detection rate at later sessions in the unassisted group was significantly higher than the detection rate during early sessions.
The trial for the study was conducted by researchers at Renmin Hospital of Wuhan University in Wuhan, China. It ran from June 2019 through October 2020.
The study defined early sessions as any procedure commencing before 1 p.m., with 58.48 percent of the procedures included in the study occurring during this time frame. Of those procedures, 357 were randomized in the unassisted group and 684 were in the AI group. A total of 739 procedures were conducted in the late session, with 263 randomized in the unassisted group and 476 in the AI group.
The study found that among the unassisted procedures, adenoma detection rate was 13.73 percent in the early sessions and 5.7 percent in the late sessions. Meanwhile, there was no significant difference in adenoma detection rates between early and later sessions with the assistance of AI systems.