Multi-center Validation of a Deep Learning Based Bowel Preparation Evaluation System

Validation of a Deep Learning Based Bowel Preparation Evaluation System: A Prospective, Multi-center, Cross-sectional Study

A deep learning based system to calculate the proportion of Boston Bowel Prep Scale (BBPS) score of 0-1 during withdrawal phase has been constructed previously. This multi-center study is going to perform a prospective observational study to validate the threshold of the adequate proportion.

Inadequate bowel preparation is insufficient for identification of polyps greater than 5 mm. However, bowel preparation assessment involved subjectivity and uncertainty. We constructed a deep learning based system to calculate the proportion of Boston Bowel Prep Scale (BBPS) score of 0-1 during withdrawal phase and performed a prospective observational study to validate the threshold of the adequate proportion.The multi-center study is aimed to verify the extrapolation and robustness of the scoring threshold based on artificial intelligence intestinal cleanliness evaluation system explored in the early stage, and propose a more accurate and quantifiable threshold for evaluating the eligibility of intestinal preparation.

Inclusion Criteria:

• Male or female aged 18 years or above;
• Ability to read, understand and sign informed consent forms;
• The researchers believe that the subjects can understand the process of the clinical study and are willing and able to complete all the study procedures and follow-up visits to cooperate with the study procedures.

Exclusion Criteria:

• Patients with contraindications to colonoscopy (obstruction or perforation, severe acute inflammatory bowel disease, toxic megacolon, severe heart failure, severe heart failure, etc.);
• Patients with galactosemia;
• Pregnant or lactating women;
• Patients used lactulose, a stimulant, or laxative within 7 days;
• Patients refused to sign informed consent forms.