An artificial intelligence (AI) tool that runs on a smartphone can help patients scheduled for a colonoscopy evaluate independently how well they do with bowel cleansing and may be an alternative approach for evaluating bowel preparation quality before the colonoscopy, especially in the COVID-19 era.
The AI tool is a “manpower-saving” option that reduces the need for nurses to evaluate the quality of bowel preparation, say Wei Gong, MD, Southern Medical University, Shenzhen, China, and colleagues.
Having the tool on a patient’s smartphone means caregivers and nurses would not be required to assess the adequacy of bowel cleansing for patients, which, in turn, would reduce person-to-person contact and the spread of infectious diseases, they add.
The study was published online in the American Journal of Gastroenterology.
Better than do-it-yourself evaluation?
The study was conducted at two hospitals in China among consecutive patients prepping for colonoscopy. All participants received standard bowel preparation instructions and were given a leaflet with general guidelines on bowel preparation.
The leaflet included photos representing bowel preparation quality and informed patients that their stool should eventually be a yellowish clear liquid; if any cloudiness (including turbid liquid, particles, or small amounts of feces) is observed in the liquid stool, the bowel preparation is not complete.
All patients were prescribed standard polyethylene glycol electrolyte solution for bowel cleansing 4-6 hours before the colonoscopy.
After consuming the solution, all patients scanned a QR (quick response) code with a smartphone for randomization into an experimental group using the AI-convolutional neural network (AI-CNN) model or a control group using self-evaluation.
The system gave instructions for using the application, taking photos of their feces, and uploading the images.
After uploading the images, the 730 patients in the AI-CNN group automatically received a “pass” or “not pass” alert, which indicated whether their bowel preparation was adequate or not.
The 704 patients in the control group evaluated the adequacy of bowel preparation on their own according to the leaflet instructions after uploading their images.
Colonoscopists and nurses were blinded to the bowel evaluation method that each patient used.
According to the investigators, evaluation results (“pass” or “not pass”) in terms of adequacy of bowel preparation as represented by Boston Bowel Preparation Scale (BBPS) scores were consistent between the two methods (AI-CNN or self-evaluation).
Overall, there were no significant differences in the two methods in terms of mean BBPS scores, polyp detection rate, or adenoma detection rate.
In subgroup analysis, however, the mean BBPS score of patients with “pass” results was significantly higher in the AI-CNN group than in the self-evaluation control group.
This suggests that the AI-CNN model may further improve the quality of bowel preparation in patients exhibiting adequate bowel preparation, the researchers say.
The results also suggest improved bowel preparation quality of the right colon under the aid of the AI-CNN model, which may be crucial for the prevention of interval colorectal cancer.
The study did not investigate the user acceptability of the AI-CNN model.
“To improve the model and broaden its application in routine practice, evaluating its convenience, accessibility, aspects that cause users difficulty, and user satisfaction is crucial,” the study team concludes.
The study was supported by the Xiamen Medical Health Science and Technology Project and the Xiamen Chang Gung Hospital Science Project. The authors have declared no relevant financial relationships.
A version of this article first appeared on Medscape.com.