Inflammatory bowel disease (IBD) is a group of chronic progressive gastrointestinal diseases that can recur throughout life and for which there is no cure. Biologics, the first line of treatment, are not only expensive, but also 30-50% of patients lack response to this type of drug therapy. However, there is a lack of reliable methods to predict the clinical efficacy of biologics. The aim of this study is to construct a reliable model to predict patients' response to biologics therapy by comprehensively analysing patients' clinical characteristics, biomarkers and other information to guide individualized therapy. In this study, we will collect clinical data from IBD patients at Peking University First Hospital, including but not limited to patients' baseline characteristics, biomarker levels, and efficacy responses, etc. We will establish a complete patient dataset using a cohort study, train and test the dataset by applying correlation analysis, multiple regression analysis, and machine learning algorithms (e.g. neural networks etc), establish and compare the prediction effects of different models The optimal model is selected for encapsulation and developed into a user-friendly clinical prediction tool. This study will contribute to clinical decision making for IBD patients, improve treatment outcomes, and reduce unnecessary healthcare costs.
Patients who are clinically diagnosed with IBD in our hospital
Disease activity is in the active stage
Receiving biologics after diagnosis
Have relevant data on clinical evaluation results after receiving biologics treatment
IBD patients who have not been treated with biologics after diagnosis
Patients diagnosed with IBD but also with other autoimmune diseases that require biologics to be treated
Patients with hematological diseases and other systemic chronic inflammatory diseases involving the intestine Patients with serious complications including malignant tumors and gastrointestinal perforation.