[Clinical research of a continuous auscultation recorder based on artificial intelligence]

Objective: To investigate the feasibility and clinical significance of a continuous auscultation recorder of bowel sounds based on artificial intelligence in monitoring the bowel sounds. Methods: From November 1,2018 to August 12,2019, a continuous auscultation recorder of bowel sounds was applied to monitor the perioperative bowel sounds of 31 patients undergoing colorectal surgery, in order to discovery underlying rules which might be used to guide clinical practice. Results: After the operation, the bowel sounds continued to exist for (1.8±0.8) h, and then gradually weakened or disappeared, and recovered gradually after (11.2±3.5) h. The first exhaust and the first defecation were detected at the time of (22.7±5.8) h and (28.7±6.9) h after surgery, respectively. The bowel sounds rate increased after eating, and decreased significantly after exhaust/defecation. Conclusions: The continuous auscultation recorder of bowel sounds based on artificial intelligence was safe and effective, which can afford help to clinical evaluation.