A synthetic intelligence (AI) device working on a smartphone can assist sufferers scheduled for a colonoscopy assess for themselves how properly they’re doing with bowel cleaning, and could also be another method to evaluate the standard of bowel preparation earlier than colonoscopy, particularly within the period of COVID-19.
The AI ​​device is a “labor-saving” choice that reduces the necessity for nurses to evaluate bowel prep high quality, say Wei Gong, MD, Southern Medical College, Shenzhen, China, and colleagues.
Having the device on a affected person’s smartphone means caregivers and nurses wouldn’t be required to evaluate the adequacy of sufferers’ bowel cleaning, which, in flip, would cut back person-to-person contact and the unfold of illness. infectious ailments, they add.
the examine was Posted on-line within the American journal of gastroenterology.
Higher than a self-assessment?
The examine was performed in two hospitals in China amongst consecutive sufferers making ready for a colonoscopy. All contributors acquired customary bowel preparation directions and got a booklet with common pointers on bowel preparation.
The bundle insert included images depicting the standard of the bowel preparation and knowledgeable sufferers that their stool ought to ultimately be a transparent, yellowish liquid; if something cloudy (together with cloudy fluid, particles, or small quantities of stool) is seen within the liquid stool, the bowel preparation just isn’t full.
All sufferers had been prescribed an ordinary polyethylene glycol electrolyte answer for bowel cleaning 4 to six hours earlier than colonoscopy.
After consuming the answer, all sufferers scanned a QR (fast response) code with a smartphone for randomization into an experimental group utilizing the AI ​​convolutional neural community mannequin (AI-CNN) or a management group utilizing self-assessment.
The system gave directions to make use of the app, take images of their stool, and add the pictures.
After importing the pictures, the 730 sufferers within the AI-CNN group mechanically acquired a “move” or “fail” alert, indicating whether or not or not their bowel prep was enough.
The 704 sufferers within the management group assessed the adequacy of the bowel preparation for themselves in response to the directions within the brochure after importing their photographs.
The colonoscopists and nurses had been unaware of the strategy of bowel analysis that every affected person used.
In line with the investigators, evaluation outcomes (“move” or “fail”) by way of adequacy of bowel preparation, as represented by Boston Bowel Preparation Scale (BBPS) scores, had been constant throughout two strategies (AI-CNN or self-assessment) .
Total, there have been no important variations between the 2 strategies in imply BBPS scores, polyp detection fee, or adenoma detection fee.
Nevertheless, in subgroup evaluation, the imply BBPS rating of sufferers with “move” outcomes was considerably increased within the AI-CNN group than within the self-assessed management group.
This implies that the AI-CNN mannequin could additional enhance bowel preparation high quality in sufferers who exhibit enough bowel preparation, the researchers say.
The outcomes additionally recommend higher bowel preparation high quality of the suitable colon with the assistance of the AI-CNN mannequin, which can be essential for hole prevention. colorectal most cancers.
The examine didn’t examine the consumer acceptability of the AI-CNN mannequin.
“So as to enhance the mannequin and increase its utility in routine follow, it’s essential to evaluate its comfort, accessibility, elements that trigger difficulties for customers, and consumer satisfaction,” the examine workforce concludes.
I’m J Gastroenterol. Posted on-line July 4, 2022. Abstract
The examine was supported by the Xiamen Medical Well being Science and Know-how Undertaking and the Xiamen Chang Gung Hospital Science Undertaking. The authors have declared no related monetary conflicts of curiosity.
For extra information, comply with Medscape on Fb, Twitter, Instagram, YoutubeY LinkedIn.