This study has three main purposes:screening: the first purpose is to evaluate the diagnostic value of combintion of the life risk factors and immunochemical fecal occult blood test (FIT) on detection of colorectal neoplasia in Chinese population; resection: the second objective is to investigate the complete resection rate of colorectal adenoma and risk factors of incomplete resection in China; identification and classification: the third objective is to initially establish an artificial intelegence-assissted recognition and classification system of polyp based on deep learning.
This study is a multi-center cross-sectional survey and diagnostic test led by the National Clinical Research Center for Digestive Disease (Shanghai) (Department of Gastroenterology, Changhai Hospital, Naval Medical University), which is conducted at about 175 digestive endoscopy centers nationwide in China, with the expectation of including 12,000 patients (10,000 screenig and 2,000 adenoma resection). The basic characteristics of patients, bowel preparation method and quality, and related information of colonoscopy are recorded in detail. According to the research purpose, the whole project can be divided into three sections. 1. Screening section: All patients receive FIT test and colonoscopy, whose age, sex, family history, smoking history, body mass index (BMI), diabetes and other risk factors are collected by researchers through pad, equipped with a specially designed database and app. Using colonoscopy results as the gold standard, the risk prediction model for the Chinese population is explored, and the optimal strategy of colonoscopy practice for the Chinese established initially. 2. Resection section: During the polypectomy, for all pathologically confirmed or NBI-predicted adenomas with size\<10mm, 1-2 biopsies were randomly performed on the edge after resection to determine the completion rate of the polypectomy. 3. Identification and classification section: For Patients regardless of cancer diagnosis or polypectomy, if there is polyp, observation of narrow band imaging (NBI) with or without magnification is required, with 4 white light and NBI images collected and reserved, respectively. If there is magnifying endoscopy, another 4 endoscopic images of magnification are also required. Endoscopists are invited to predict the pathology of polyps according to the NBI International Colorectal Endoscopic (NICE) classification principle and endoscopic images, and upload the pathological results and endoscopic images within 2-4 week after colonoscopy.
Study Type
OBSERVATIONAL
Enrollment
12,000
All included patients received FIT test and then colonoscopy, with the risk factors of CRC recorded. The diagnostic performance of predicting model (based on FIT and risk factors) and colonoscopy were compared.
All included patients received polypectomy, and then biopsy is performed on the edge of resection for patients with \< 10 mm adenoma (confirmed by pathology or predicted by NBI images), with the complete resection rate of polyps being calculated.
Pathology of polyps is classified by endoscopists through NICE principle and the performance of classification between endoscopists and computer is compared.
Changhai Hospital, Second Military Medical University
Shanghai, China
Detection of colorectal neoplasia
Colorectal neoplasia included CRCs, adenomas, sessile serrated adenomas/polyps, traditional serrated adenomas/polyps, and hyperplastic polyps ≥10 mm, which were recommended to have a shorter surveillance interval after polypectomy
Time frame: 24 hours
Complete resection rate of polypectomy
Time frame: 24 hours
Specificity and sensitivity of endoscopists and artificial intelligence-assisted system in classifying polyps
Time frame: 24 hours
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.