Lee Alfred Lok Hang, To Curtis Chun Kit, Chan Ronald Cheong Kin, Wong Janus Siu Him, Lui Grace Chung Yan, Cheung Ingrid Yu Ying, Chow Viola Chi Ying, Lai Christopher Koon Chi, Ip Margaret, Lai Raymond Wai Man
Department of Microbiology, Prince of Wales Hospital, Shatin, Hong Kong SAR, China.
Department of Anatomical and Cellular Pathology, Faculty of Medicine, Chinese University of Hong Kong, Shatin, Hong Kong SAR, China.
JAC Antimicrob Resist. 2024 Aug 7;6(4):dlae121. doi: 10.1093/jacamr/dlae121. eCollection 2024 Aug.
To develop an artificial intelligence model to predict an antimicrobial susceptibility pattern in patients with urinary tract infection (UTI).
26 087 adult patients with culture-proven UTI during 2015-2020 from a university teaching hospital and three community hospitals in Hong Kong were included. Cases with asymptomatic bacteriuria (absence of diagnosis code of UTI, or absence of leucocytes in urine microscopy) were excluded. Patients from 2015 to 2019 were included in the training set, while patients from the year 2020 were included as the test set.Three first-line antibiotics were chosen for prediction of susceptibility in the bacterial isolates causing UTI: namely nitrofurantoin, ciprofloxacin and amoxicillin-clavulanate. Baseline epidemiological factors, previous antimicrobial consumption, medical history and previous culture results were included as features. Logistic regression and random forest were applied to the dataset. Models were evaluated by F1-score and area under the curve-receiver operating characteristic (AUC-ROC).
Random forest was the best algorithm in predicting susceptibility of the three antibiotics (nitrofurantoin, amoxicillin-clavulanate and ciprofloxacin). The AUC-ROC values were 0.941, 0.939 and 0.937, respectively. The F1 scores were 0.938, 0.928 and 0.906 respectively.
Random forest model may aid judicious empirical antibiotics use in UTI. Given the reasonable performance and accuracy, these accurate models may aid clinicians in choosing between different first-line antibiotics for UTI.
开发一种人工智能模型，以预测尿路感染（UTI）患者的抗菌药物敏感性模式。
纳入了2015年至2020年期间来自香港一所大学教学医院和三家社区医院的26087例经培养证实为UTI的成年患者。排除无症状菌尿症患者（无UTI诊断代码，或尿显微镜检查无白细胞）。2015年至2019年的患者纳入训练集，2020年的患者纳入测试集。选择三种一线抗生素来预测引起UTI的细菌分离株的敏感性：即呋喃妥因、环丙沙星和阿莫西林-克拉维酸。将基线流行病学因素、既往抗菌药物使用情况、病史和既往培养结果作为特征纳入。将逻辑回归和随机森林应用于数据集。通过F1分数和曲线下面积-受试者操作特征（AUC-ROC）对模型进行评估。
随机森林是预测三种抗生素（呋喃妥因、阿莫西林-克拉维酸和环丙沙星）敏感性的最佳算法。AUC-ROC值分别为0.941、0.939和0.937。F1分数分别为0.938、0.928和0.906。
随机森林模型可能有助于在UTI中明智地使用经验性抗生素。鉴于其合理的性能和准确性，这些准确的模型可能有助于临床医生在为UTI选择不同的一线抗生素之间做出决策。
