AUTHOR=Wang Zhiyu , Ong Chiat Ling Jasmine , Fu Zhiyan TITLE=AI Models to Assist Vancomycin Dosage Titration JOURNAL=Frontiers in Pharmacology VOLUME=Volume 13 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2022.801928 DOI=10.3389/fphar.2022.801928 ISSN=1663-9812 ABSTRACT=Background: Effective treatment using antibiotic Vancomycin requires close monitoring of serum drug levels due to its narrow therapeutic index. In current practice, physicians use various dosing algorithms for dosage titration, but these algorithms reported low success in achieving therapeutic target. We explored using artificial intelligent (AI) to assist vancomycin dosage titration. Methods: We used a novel method to generate the label for each record, and only included records with appropriate label data to generate a clean cohort with 2,282 patients and 7,912 injection records. Among them, 64% of patients were used to train two machine learning models, one for initial dose recommendation and one for subsequence dose recommendation. The model performance was evaluated using two metrics: PAR, a pharmacology meaningful metric defined by us, and MAE, a commonly used regression metric. Results: In our three year’s data, only small portion (34.1%) of current injection doses could reach the desired vancomycin trough level (14-20 mcg/ml). Both PAR and MAE of our machine learning models were better than the classical pharmacokinetic models. In some cases when the recommendation of our model showed larger different from the label, our recommendation might offer an alternative titration pathway to achieve the therapeutic effect. Conclusions: We developed machine learning models to recommend vancomycin dosage. Our results showed that the new AI-assisted dosage titration approach have the potential to improve the traditional approaches. This is especially useful to guide decision making for inexperienced doctors in making consistent and safe dosing recommendations for high-risk medications like vancomycin.