Erratum to “Translation of Machine Learning-Based Prediction Algorithms to Personalised Empiric Antibiotic Selection: A Population-Based Cohort Study” International Journal of Antimicrobial Agents 62 (2023) 106966 (International Journal of Antimicrobial Agents (2023) 62(5), (S0924857923002455), (10.1016/j.ijantimicag.2023.106966))

The publisher regrets error that occurred in the figure legends of this article. The error pertains to Figures 2–4 and those legends are inappropriately matched. The corrected information is below. Figure 2. Variable importance based on the absolute Shapley additive explanations (SHAP) of two different prediction models (the ciprofloxacin and piperacillin/tazobactam models). SHAP is a model-agnostic tool that assesses importance through a predictor's contribution to a given prediction compared with the average prediction. Abbreviations: LLR, Lasso logistic regression; RF, random forest; XGB, extreme gradient boosting machine; MDRO, multidrug-resistant organism, IV, intravenous; PO, per os; AMP, ampicillin; AMS, ampicillin/sulbactam; CFP, cefepime; CIP, ciprofloxacin; GEN, gentamicin; PPT, piperacillin/tazobactam; SXT, sulfamethoxazole/trimethoprim; IPM, imipenem; HT, hypertension; DM, diabetes mellitus; CKD, chronic kidney disease; ICU, intensive care unit; VRE, vancomycin-resistance enterococci; MRSA, methicillin resistant Staphylococcus aureus; CRE, carbapenem-resistant Enterobacteriaceae; ESBL, extended-spectrum beta-lactamase producing Enterobacteriaceae; PC, penicillin; Cefa or Cepha, cephalosporin; FQ, fluoroquinolone. Figure 3. Shiny application including patient clinical characteristics input and calculated results for probabilities of predicted non-susceptibility. The results of individual antibiotic susceptibility tests were predicted through clinical information. Abbreviations: MDRO, multidrug-resistant organism; IV, intravenous; PO, per os; AMP, ampicillin; AMS, ampicillin/sulbactam; CFP, cefepime; CIP, ciprofloxacin; GEN, gentamicin; PPT, piperacillin/tazobactam; SXT, sulfamethoxazole/trimethoprim; IPM, imipenem; HT, hypertension; DM, diabetes mellitus; CKD, chronic kidney disease; ICU, intensive care unit; VRE, vancomycin-resistance enterococci; MRSA, methicillin-resistant Staphylococcus aureus; CRE, carbapenem-resistant Enterobacteriaceae; ESBL, extended-spectrum beta-lactamase producing Enterobacteriaceae; PC, penicillin; Cefa, cephalosporin; FQ, fluoroquinolone. Figure 4. Predicted non-susceptibility rate for various patients’ scenarios. From left to right, the characteristics corresponding to the column are accumulated. (A) Baseline case with 40-year-old female patient hospitalised on 25 December 2018; (B) 40-year-old male patient hospitalised on the same date; (C) 70-year-old female patient hospitalised on the same date; (D) 40-year-old female patient hospitalised on 7 February 2011. Abbreviations: MDRO, multidrug-resistant organism; abx, antibiotics; IV, intravenous; PO, per os; AMP, ampicillin; AMS, ampicillin/sulbactam; CFP, cefepime; CIP, ciprofloxacin; GEN, gentamicin; PPT, piperacillin/tazobactam; SXT, sulfamethoxazole/trimethoprim; IPM, imipenem; HT, hypertension; DM, diabetes mellitus; CKD, chronic kidney disease; ICU, intensive care unit; VRE, vancomycin-resistance enterococci; MRSA, methicillin-resistant Staphylococcus aureus; CRE, carbapenem-resistant Enterobacteriaceae; ESBL, extended spectrum beta-lactamase producing Enterobacteriaceae; PC, penicillin; Cefa, cephalosporin; FQ, fluoroquinolone. The publisher would like to apologise for any inconvenience caused.