In this study, the investigators proposed an artificial intelligence-based preoperative automatic reminder system on colonocopy, which can improve the quality of bowel preparation and the rate of polyps and adenomas detection.

Despite advances in bowel preparation methods, bowel preparation is inadequate in up to one-third of all colonoscopies in reported series. Inadequate bowel cleansing results in negative con-sequences for the examination, including incomplete visualization of the colon, missed lesions(22-48%), procedural difficulties, prolonged procedure time and reduced time interval until follow-up, and an estimated 12-22% increase in overall colonoscopy cost.

The adequacy of a bowel preparation is closely linked to patient compliance with both dietary and purge instructions. Previous work has shown that 18-23.5% of the patients with poor preparation had failed to follow preparation instructions. One study performed in Asia showed that non-compliance with bowel preparation instructions, lower education level, and a long wait for the colonoscopy appointment were independent risk factors for poor bowel preparation. A survey among doctors showed that gastroenterologists with the highest number of patients with inadequate bowel preparation believed that patients are unwilling to follow preparation instructions, struggle with the prescribed diet, and are unable to tolerate the full course of purgative. It is reasonable to hypothesis that efforts to improve education and maximize patient compliance during the preparatory period will enhance the efficacy of bowel preparation.

A research has shown that telephone-based re-education about the details of bowel preparation on the day before colonoscopy significantly improved the quality of bowel preparation and polys detection rate. In recent years, artificial intelligence (AI) has been successfully applied in multiple medical fields. But there has not been an artificial-intelligence-based system which can automatically remind patients of the details of bowel preparation on the day before colonoscopy.

In this study, we proposed an artificial intelligence-based preoperative automatic reminder system on colonoscopy, which can improve the quality of bowel preparation and the rate of polyps and adenomas detection.

Inclusion Criteria:

1. Male or female between 50-75 years old;
2. Able to read, understand and sign informed consent
3. The investigator believes that the subjects can understand the process of the clinical study, are willing and able to complete all study procedures and follow-up visits, and cooperate with the study procedures
4. Outpatients requiring colonoscopy

Exclusion Criteria:

1. colorectal surgical history
2. Severe colonic stricture or obstructive tumor
3. Obvious gastroparesis or gastric outlet obstruction or intestinal infarction
4. Pregnant or lactating women
5. Dysphagia or impaired swallowing reflex or mental state
6. Intestinal obstruction or perforation is known
7. Severe chronic renal failure（CCR< 30ml/min）
8. Severe congestive heart failure（NYHA ⅢorⅣ）
9. Uncontrolled hypertension( systolic pressure>170mmHg，diastolic pressure>100mmHg)
10. Dehydration or electrolyte disorder
11. Toxic colitis or megacolon
12. Inable to give informed consent
13. Hemodynamic instability