Detection and categorization of colonoscopy images of Ulcerative Colitis (UC) patients using Artificial Intelligence (AI)

The study will begin by obtaining approval from the Institutional Ethics Committee of Kasturba Medical College, MAHE, Manipal. Ethical clearance ensures that the research follows ethical guidelines, prioritizing participant safety and data privacy. Participants will be fully informed about the study’s purpose, risks, and benefits, and written informed consent will be obtained before any procedures commence.

Recruitment Process

Participants will be recruited from the outpatient and inpatient departments of the Gastroenterology and Hepatology department at Kasturba Medical College. The recruitment will focus on adults diagnosed with Ulcerative Colitis (UC), with participants selected based on predefined inclusion and exclusion criteria. Screening will involve reviewing medical histories, clinical records, and imaging studies to confirm the UC diagnosis. This recruitment process will be conducted over a set period to ensure a representative and adequate sample size for the study.

Consenting Process

After identifying potential participants, they will receive detailed information about the study, including its objectives, procedures, potential risks, and benefits. Informed consent will be obtained before any study-related assessments begin, and participants will be informed of their right to withdraw from the study at any time without affecting their standard medical care.

Assessments and Data Collection

Upon recruitment, participants will undergo an initial assessment, which includes collecting demographic data (such as age and gender) and conducting a clinical evaluation. Laboratory investigations (e.g., complete blood count, C-reactive protein, and fecal calprotectin) will be performed to assess disease activity. The key diagnostic procedure in the study is a colonoscopy/sigmoidoscopy, during which images will be captured for AI analysis. These endoscopic images will be scored using the Mayo Endoscopic Score (MES) to assess the activity of UC.

The Mayo Endoscopic Score (MES) is a straightforward and reliable tool used to gauge the level of inflammation in the colon of people with ulcerative colitis (UC). It’s based on what doctors observe during a colonoscopy and is a key part of the Mayo Score, which is commonly used in both medical practice and research.

The MES breaks down the appearance of the colon into four categories:

·       Grade 0: The colon looks normal or shows no signs of active inflammation.

·       Grade 1: There are mild changes, such as slight redness, a reduced vascular pattern, and some mild fragility in the mucosa.

·     Grade 2: Moderate inflammation is evident, with pronounced redness, the loss of visible blood vessels, fragility, and small erosions.

·       Grade 3: The inflammation is severe, with ulcers and spontaneous bleeding.

The collected images will undergo preprocessing to ensure they are standardized for AI analysis. This preprocessing involves adjusting brightness, contrast, and removing noise. Along with these images, clinical data will be integrated to help in building a personalized AI model that can assist in detecting and categorizing UC according to MES.

Intervention

This is not an interventional study in terms of treatment. However, the AI diagnostic tool will act as an intervention by categorizing the severity of UC based on colonoscopy/sigmoidoscopy images. The AI system will complement the standard diagnostic approach by providing an automated, objective assessment of disease severity. No additional procedures beyond standard care will be imposed on participants.

Study Endpoint and Withdrawal Criteria

The study’s endpoint will focus on successfully developing an AI model capable of detecting and categorizing UC severity. Participants can withdraw from the study at any point, and their data will be anonymized. Withdrawal will not affect their access to clinical care or ongoing medical treatment.