IDDF2024-ABS-0127 A predictive machine-learning model on clinical decision-making for washed microbiota transplantation in treatment of ulcerative colitis

<h3>Background</h3> Machine learning based on clinical data and treatment protocols for better clinical decision-making is a current research hotspot. This study aimed to build a machine learning model on washed microbiota transplantation (WMT) for ulcerative colitis (UC), providing patients and clinicians with a new evaluation system to optimize clinical decision-making. <h3>Methods</h3> Patients with UC who underwent WMT via mid-gut or colonic delivery route at an affiliated hospital of Nanjing Medical University from April 2013 to June 2022 were recruited. Model ensembles based on the clinical indicators were constructed by machine-learning to predict the clinical response of WMT after one month. <h3>Results</h3> A total of 366 patients were enrolled in this study, with 210 patients allocated for training and internal validation and 156 patients for external validation. The low level of indirect bilirubin, activates antithrombin III, defecation frequency, and cholinesterase and the high level of age, creatine kinase, HCO₃^- and thrombin time were related to the clinical response of WMT at one month. Besides, the voting ensembles exhibited an area under curve (AUC) of the voting ensembles was 0.769 ± 0.019 [accuracy, 0.754; F1-score, 0.845] in the internal validation; the AUC of the external validation was 0.614 ± 0.017 [accuracy, 0.8013; F1-score, 0.887]. <h3>Conclusions</h3> This study first conducted a machine learning model with the largest sample size to predict the clinical response of WMT on UC at one month. This study also implies machine learning has potential value in the field of WMT.