Skip to Content
  • Proceeding Paper
  • Open Access

28 August 2026

8 Pages

The Future Step in Antibiotic Stewardship: Is Artificial Intelligence a Real Solution? †

and
1
Faculty of Medicine and Biological Science, Stefan cel Mare University of Suceava, 720229 Suceava, Romania
2
Emergency Clinical Hospital New St. John Suceava, 720224 Suceava, Romania
*
Author to whom correspondence should be addressed.
†
Presented at the International Conference on Electromagnetic Fields, Signals and BioMedical Engineering (ICEMS-BIOMED), Suceava, Romania, 7–9 May 2026.

Abstract

The problem of antimicrobial resistance (AMR) is a significant worldwide health risk, which is why experts are interested in using artificial intelligence (AI) or machine learning (ML) to employ antibiotics more wisely. Diagnostic, empiric antibiotic choices, dosing, and surveillance in stewardship programs are AI-driven promises to make the process straightforward. This narrative review is a synthesis of the recent evidence (within the last 10 years) on the use of AI in antibiotic stewardship. It analyzed the literature on AI-based diagnostic systems, prescribing decision-support systems, resistance prediction models, real-time surveillance systems, and dose algorithms. The use of AI tools by clinicians and stewardship programs must be approached with discretion and human supervision and assessment.

1. Introduction

The issue of antimicrobial resistance (AMR) is a worldwide problem, with millions of deaths each year being related to drug-resistant infections [1,2]. Conventional stewardship practices are based on culture-oriented diagnostics and diagnostic experts, which tend to result in delayed or excessive broad empiric therapy [3]. The opportunities brought by artificial intelligence (AI) and machine learning (ML) are due to the analysis of large, complex healthcare datasets to inform the use of antibiotics [4]. As an example, using ML algorithms, it becomes possible to analyze the laboratory or genomic data quickly and identify the pathogens and resistance genes, or comb through the electronic health records (EHRs) to determine which antibiotic is most likely to work [5]. In antibiotic stewardship programs (ASPs), AI may be used to aid activities like flagging inappropriate prescriptions, predicting organ-specific risks of resistance, and automating aspects of audit-and-feedback (e.g., by identifying hotspots of unnecessary broad-spectrum use) [6]. The latest reviews highlight the potential of AI in diagnostic, surveillance, and therapeutic decision-making. Genomic datasets can be used to reveal new patterns of resistance with the help of AI-enhanced surveillance systems [7]. In diagnostics, deep learning has been utilized including rapid pathogen identification (ID) and syndrome classification (e.g., differentiating between bacterial and viral infections). Notably, AI tools in ASPs are meant to support (not to substitute) clinical judgment [8,9]. This review analyzes the existing data on AI in antibiotic stewardship, specifically its application in clinical settings, reported advantages, limitations/challenges, and considerations of its implementation. The primary research and systematic reviews during 2010–2026 were summarized, pointing out such examples and outlining the development of the field.

2. Materials and Methods

A targeted literature search was conducted in PubMed, PubMed Central, Web of Science and other databases for peer-reviewed studies (≥2015) and reviews on “AI or machine learning” and “antibiotic or antimicrobial stewardship.” Keywords included “AI antibiotic stewardship,” “machine learning resistance prediction,” “decision support antibiotic,” “AI dosing antibiotics,” etc. Only studies with full-text availability (e.g., observational studies, quantitative studies, systematic reviews, and meta-analyses) were considered. Additionally, publications such as editorials, letters, author responses, and review articles were excluded, as were studies without accessible full texts. Emphasis was placed on recent systematic reviews, narrative reviews, and key primary studies. Relevant references from core articles were also scanned. Studies were categorized by application (diagnostics, prescribing decision support, resistance prediction, surveillance, dosing, and workflow automation).

3. AI Technologies Relevant to Stewardship

Machine learning (ML) is an algorithm-based approach (e.g., decision trees, random forests, gradient boosting, neural networks) to identify patterns in large datasets. Deep learning (DL) (deep neural networks, CNNs) is particularly good at feature extraction from complex data such as genomics and images. ML and DL have already found applications in healthcare, where they assist in diagnosis (e.g., pneumonia detection on X-rays), genomic analyses, and personalized therapy prediction [10]. ML models can be used in stewardship to forecast antimicrobial resistance or optimal empiric decisions based on patient factors, local epidemiology, and the microbiology history. As an example, ML-based patient-specific resistance predictions in the form of antimicrobial-specific antibiograms have been trained to predict resistance in specific patients and across major antibiotics [11,12]. Neural networks can also learn trends in complex data (e.g., ICU time-series) to inform dosing changes or sepsis. Natural Language Processing (NLP) techniques can be used to extract pertinent information in clinical notes and guidelines to aid in decision-making, and Large Language Models (LLMs) (e.g., GPT-4, BERT) are able to provide a summary of the literature or suggest treatment recommendations based on textual queries. Early research suggests that LLMs generally fail at answering open-ended infectious disease queries; ChatGPT has been shown to make more than 30% errors on complicated cases [13]. The DeepSeek models are comparable to those of ChatGPT, with DeepSeek-V3 outperforming it in certain performance metrics [14]. LLMs can offer general knowledge or education, but the existing evidence indicates that they do not have any reliable, evidence-based clinical accuracy when not overseen by experts. Proper engineering in a timely manner can improve this, yet LLMs are not tested on frontline stewardship decisions [15]. Clinical Decision Support Systems (CDSSs) are AI-based systems integrated into electronic health records (EHRs) or ordering systems to provide real-time advice (e.g., highlighting narrower-spectrum agents or notifying professionals of IV-to-oral switching opportunities) [16]. Predictive analytics is the application of patient data to predict (e.g., risk of resistant infection). As one example, ML algorithms could be used to predict inpatients with the likelihood of contracting drug-resistant pathogens, which trigger more focused empiric therapy [17]. It is also possible to combine patient microbiology (e.g., blood culture results) and clinical data with AI to recommend the optimal antibiotic without final sensitivities [18]. AI (convolutional neural network—CNN) has the potential to differentiate between bacterial and viral infections on a chest radiograph or CT scan and, therefore, potentially lower the use of unnecessary antibiotics in viral pneumonia [19,20]. Although imaging AI is yet to find substantial application in stewardship, proof-of-concept experiments indicate that DL can recognize features (e.g., pneumonia patterns) more quickly and reliably than humans. This helps indirectly in stewardship by enlightening clinicians about the situations in which they are unlikely to need antibiotics. AI can reduce time to identification and resistance profiles by many factors, with rapid pathogen sequencing [21]. Antibiotic susceptibilities can be predicted with the use of ML models that run on genomic or transcriptomic data. As an illustration, merging genomic data with ML (like in the system used by Arkstone) provided accurate, consensus-driven recommendations in just a few hours (n = 17–29)—orders of magnitude quicker than standard culture susceptibility [22]. This can be done by using such tools to speed up the de-escalation of broad antibiotics. AI can also be used to analyze host and microbiome genomics to forecast infection risk or tailor antibiotic dosage in the future [23]. In addition to radiology, AI used in new diagnostics (e.g., blood-based markers, point-of-care tests) may identify bacterial vs. non-bacterial infection in minutes, informing the need to use antibiotics [24]. Microscopy images (e.g., Gram stain or urine sediment) can also be further automated using deep learning, which can be used to further automate rapid diagnostics [25]. In short, AI involves various technologies, including the older models based on ML classification/regression and the newer models based on LLMs, which are being considered to assist with smarter antimicrobial use. Initial uses include ML-based decision support (resistance prediction and optimal therapy), whereas the tools based on LLMs are still under experimentation in stewardship.

4. AI vs. Traditional Stewardship Approaches

Conventional ASPs (audit/feedback, preauthorization) can generally decrease inappropriate utilization and quantity of antibiotics by up to 20–30 percent in intervention wards [26,27,28]. Similar or even larger effects have been demonstrated by AI-augmented CDSSs. The INSPIRE trials reported further 11–28% relative decreases in extended-spectrum antibiotic days beyond the baseline stewardship [29,30,31]. One of the predecessors of modern AI was the Clinical Decision Support (CDS), which reduced the confirmed diagnosis time and hospitalization days [32]. These results indicate that ML-based interventions may be enhanced when overlayed with routine stewardship. Guidelines and patient data are directly integrated into AI tools, and frequently have high concordance with expert recommendations [33]. Conversely, manual audit is based on human examination, and AI is able to use the rule in a consistent and alert–fatigued manner. No population-level decrease in resistance due to AI stewardship has been demonstrated as it needs long-term massive surveillance. Nevertheless, AI tools ought to contribute to decreasing resistance development by cutting down on unneeded broad-spectrum consumption. Conventional ASPs have been associated with the stabilization of or the reduction in resistance among hospitals. The quantitative effect of AI is yet to be determined. Most importantly, there is no evidence of adverse outcomes as AI-guided prescribing increases [34]. Bolton et al. discovered no patient safety measure was worsened by the introduction of an AI CDSS; clinicians were able to ignore bad advice [35]. The INSPIRE data did not indicate any increase in ICU transfers [29,30,31]. Furthermore, accelerated optimal therapy (with the help of AI) will enhance performance with severe infections (indirectly supported by sepsis AI research) [36]. Conventional stewardship also demonstrates no or a positive influences on outcomes. So far AI seems to be at least as safe as more traditional methods as long as it is properly validated. There are few formal cost-effectiveness studies of AI in stewardship. Nevertheless, the saving of broad-spectrum antibiotic days and hospital stay by even a fraction will pay off investment in AI tools. As an example, the saving of approximately 700 IV days/year would probably reduce costs and complications [4]. AI can have initial software expenditure and data infrastructure, but it can decrease clinician workload (less manual audit). The conventional programs necessitate constant personnel (pharmacists, epidemiologists). An essential gap in research is economic evaluations. The AI interventions will mostly be implemented into the current workflow (EHR alerts or dashboards), which may be more effective than manual audits that are time-consuming. According to Bolton et al., clinicians found the AI CDSS helpful (TAM scores were around 3.8/5 for ease-of-use) [35]. Nonetheless, it is important to lean on explanations and trust, as in the study, most clinicians seldom sought the rationale of the AI (9% of the decisions) and embraced the AI advice mainly when it was based on evidence and was user-friendly [35]. Cultural change is needed in traditional stewardship, but AI tools have special obstacles. To conclude, AI systems can help to improve or even speed up the old-fashioned metrics of stewardship, but direct comparisons with the traditional ASPs are not possible. They are to be regarded as complementary: risk stratification and alerts can be automated with the help of AI, whereas stewardship teams should have oversight and contextually judge.

5. Benefits of AI in Antibiotic Stewardship

AI has a number of tangible advantages for antibiotic stewardship initiatives. AI algorithms are able to reveal complicated trends and hidden risk factors that may be overlooked by human analysis. The meta-analysis study by Pinto et al. (2025) discovered that ML was more effective than conventional analyses in terms of sensitivity and negative predictive value (NPV) because fewer resistant cases were neglected [34]. Better diagnostic and predictive accuracy aids in better matching therapy to patient needs. Data are processed by automated tools, which do not become tired. As an example, a pharmacist AI rule engine can screen all inpatient orders in real time, unlike human pharmacists who cannot cover all cases [36]. By reducing unnecessary broad-spectrum prescribing (e.g., saving 28% in the INSPIRE trial) and streamlining audits (Goodman et al.’s ~50% work reduction), AI frees up stewardship resources for high-value tasks and enables truly personalized antibiograms and dosing [29,37,38]. Instead of using hospital-wide statistics, ML models can estimate risk on a personalized basis (e.g., by the co-morbidities and previous cultures of a patient) [39]. This results in safer empiric therapy. An example is the OPTIVAN vancomycin model that varies dose based on the pharmacokinetics of a particular patient [39]. AI models are able to constantly consume incoming lab results and identify outbreaks or trends in resistance earlier than manual systems [40]. Detection of increasing AMR in near real time and alerting stewardship teams enables quicker action (e.g., changing hospital formularies). Interpretable models (such as the SHAP-based model proposed by Cavallaro and the decision trees proposed by Yang) can be used to uncover why a recommendation is given [9,41]. This will not only create clinician trust but also have the potential to inform a provider about risk factors of resistance, strengthening good prescribing behavior.

6. Limitations and Challenges

Despite advantages, AI tools will have to address challenges before they can be used in a wide range of clinical applications. Numerous ML models are based on extensive retrospective data. These data are subject to fluctuations in quality: gaps in fields, coding mistakes and variations in measurement can distort model training. To illustrate this, when an EHR database is not reporting some infections, an ML model can underpredict those cases [42]. The heterogeneity of results of the studies (different hospitals, disease mix) makes it difficult to compare them directly [43]. In the absence of quality representative data (both outpatient and pediatric settings), models face the risk of low generalizability. Patterns of historical care, trained in the model, could be biased (e.g., socioeconomic or racial differences). Few studies have focused on fairness in antibiotic AI, and any implementation must ensure the absence of unequal performance among subpopulations. Black-box models (e.g., deep neural nets) may be highly accurate, but provide no information to users about the process behind how they arrive at their predictions. This opaqueness causes clinical denial and poses accountability questions. The regulatory agencies are also cautious of opaque AI as well. The integration of such technology into hospital IT systems is not easy. Even at the moment, there is a risk of alert fatigue: excessive amounts of AI warnings or improperly tuned prompts can be disregarded [44]. Effective algorithms can be of little use unless human factors are carefully designed and trained. There should be transparency, validation and compliance with medical device regulations. Privacy of data (particularly in the case of genomic or inter-institutional data-sharing) should be ensured. The creation and sustenance of AI tools need knowledge in data science, informatics infrastructure, and ongoing updates. These resources may not be available in smaller hospitals or low-resource settings. Collaborations (e.g., with academic centers) or cloud-based applications may work; however, sustainability is an issue.

7. Conclusions

AI and ML are rapidly gaining traction in antibiotic stewardship, with promising applications across the workflow from diagnosis to dosing. Studies have shown that AI systems can match or improve upon conventional care in predicting resistance and guiding therapy, while saving resources (e.g., reducing pharmacy reviews) and potentially improving patient safety. The technology enables personalized, data-driven decision-making, a major advancement over one-size-fits-all antibiograms. However, the current evidence is preliminary. Most studies are single-center or retrospective, and real-world implementation has been limited. Challenges of data quality, bias, interpretability, and integration into busy clinical workflows must be addressed before AI can be a standard stewardship tool. Ethical and regulatory frameworks are lagging behind the technology. In the near term, stewardship programs should consider pilot projects where AI augments (not replaces) human expertise. Clinicians must remain the final arbiters, using AI recommendations as one input.

Author Contributions

Conceptualization, O.A.C.-S., methodology, O.A.C.-S. and R.F.; formal analysis, O.A.C.-S. and R.F. investigation, O.A.C.-S. and R.F.; resources, O.A.C.-S. and R.F.; writing—original draft preparation, O.A.C.-S. and R.F.; writing—review and editing, O.A.C.-S. and R.F.; visualization, O.A.C.-S. and R.F. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript: AI—Artificial Intelligence; CDSS—Clinical Decision Support System; AMR = Antimicrobial Resistance; ASP—Antibiotic Stewardship Program; Clinical Decision Support—CDS; CNN—Convolutional Neural Network; CPOE—Computerized Provider Order Entry; DL—Deep Learning; ED—Emergency Department; EHR—Electronic Health Record; ICU—Intensive Care Unit; LLM—Large Language Model; MDRO—Multidrug-Resistant Organisms; ML—Machine Learning; NLP—Natural Language Processing; NPV—Negative Predictive Value.

References

  1. Pennisi, F.; Pinto, A.; Ricciardi, G.E.; Signorelli, C.; Gianfredi, V. The Role of Artificial Intelligence and Machine Learning Models in Antimicrobial Stewardship in Public Health: A Narrative Review. Antibiotics 2025, 14, 134. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  2. Arnold, A.; McLellan, S.; Stokes, J.M. How AI can help us beat AMR. npj Antimicrob. Resist. 2025, 3, 18. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  3. Corbin, C.K.; Sung, L.; Chattopadhyay, A.; Noshad, M.; Chang, A.; Deresinksi, S.; Baiocchi, M.; Chen, J.H. Personalized antibiograms for machine learning driven antibiotic selection. Commun. Med. 2022, 2, 38. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  4. Harandi, H.; Shafaati, M.; Salehi, M.; Roozbahani, M.M.; Mohammadi, K.; Akbarpour, S.; Rahimnia, R.; Hassanpour, G.; Rahmani, Y.; Seifi, A. Artificial intelligence-driven approaches in antibiotic stewardship programs and optimizing prescription practices: A systematic review. Artif. Intell. Med. 2025, 162, 103089. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  5. Anahtar, M.N.; Yang, J.H.; Kanjilal, S. Applications of Machine Learning to the Problem of Antimicrobial Resistance: An Emerging Model for Translational Research. J. Clin. Microbiol. 2021, 59, e0126020. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  6. Oliveira, M.; Antunes, W.; Mota, S.; Madureira-Carvalho, Á.; Dinis-Oliveira, R.J.; Dias da Silva, D. An Overview of the Recent Advances in Antimicrobial Resistance. Microorganisms 2024, 12, 1920. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  7. Ayesiga, I.; Yeboah, M.O.; Okoro, L.N.; Edet, E.N.; Gmanyami, J.M.; Ovye, A.; Atimango, L.; Gadzama, B.N.; Kembabazi, E.; Atwau, P. Artificial intelligence-enhanced biosurveillance for antimicrobial resistance in sub-Saharan Africa. Int. Health 2025, 17, 795–803. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  8. De la Lastra, J.M.P.; Wardell, S.J.T.; Pal, T.; de la Fuente-Nunez, C.; Pletzer, D. From Data to Decisions: Leveraging Artificial Intelligence and Machine Learning in Combating Antimicrobial Resistance—A Comprehensive Review. J. Med. Syst. 2024, 48, 71. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  9. Cavallaro, M.; Moran, E.; Collyer, B.; McCarthy, N.D.; Green, C.; Keeling, M.J. Informing antimicrobial stewardship with explainable AI. PLoS Digit. Health 2023, 2, e0000162. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  10. Shobayo, O.; Saatchi, R. Developments in Deep Learning Artificial Neural Network Techniques for Medical Image Analysis and Interpretation. Diagnostics 2025, 15, 1072. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  11. Sakagianni, A.; Koufopoulou, C.; Feretzakis, G.; Kalles, D.; Verykios, V.S.; Myrianthefs, P.; Fildisis, G. Using Machine Learning to Predict Antimicrobial Resistance-A Literature Review. Antibiotics 2023, 12, 452. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  12. Yuan, K.; Luk, A.; Wei, J.; Walker, A.S.; Zhu, T.; Eyre, D.W. Machine learning and clinician predictions of antibiotic resistance in Enterobacterales bloodstream infections. J. Infect. 2025, 90, 106388. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  13. AlGain, S.; Marra, A.R.; Kobayashi, T.; Marra, P.S.; Celeghini, P.D.; Hsieh, M.K.; Shatari, M.A.; Althagafi, S.; Alayed, M.; Ranavaya, J.I.; et al. Can we rely on artificial intelligence to guide antimicrobial therapy? A systematic literature review. Antimicrob. Steward. Healthc. Epidemiol. 2025, 5, e90. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  14. Zhang, L.; Pan, Y.; Lai, W.; Liang, Z.; Zhong, H.; Lin, X. Benchmark evaluation of deepseek AI models in antibacterial clinical decision-making for infectious diseases. BMC Med. Inf. Decis. Mak. 2026, 26, 62. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  15. Marra, A.R.; Langford, B.J.; Nori, P.; Bearman, G. Revolutionizing antimicrobial stewardship, infection prevention, and public health with artificial intelligence: The middle path. Antimicrob. Steward. Healthc. Epidemiol. 2023, 3, e219. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  16. Peiffer-Smadja, N.; Rawson, T.; Ahmad, R.; Buchard, A.; Georgiou, P.; Lescure, F.-X.; Birgand, G.; Holmes, A. Machine learning for clinical decision support in infectious diseases: A narrative review of current applications. Clin. Microbiol. Infect. 2020, 26, 584–595. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  17. De La Torre, J.C.G.; Frenkel, A.; Chavez-Lencinas, C.; Rendon, A.; Higuchi, Y.; Vela-Ruiz, J.M.; Calpey, J.; Beaton, R.; Elijah, I.; Shachar, I.; et al. Guiding Antibiotic Therapy with Machine Learning: Real-World Applications of a CDSS in Bacteremia Management. Life 2025, 15, 1756. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  18. de la Torre, J.C.G.; Frenkel, A.; Chavez-Lencinas, C.; Rendon, A.; Cáceres, J.A.; Alvarado, L.; Hueda-Zavaleta, M. AI-Based Treatment Recommendations Enhance Speed and Accuracy in Bacteremia Management: A Comparative Study of Molecular and Phenotypic Data. Life 2025, 15, 864. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  19. Assudani, P.J.; Bhurgy, A.S.; Kollem, S.; Bhurgy, B.S.; Ahmad, O.; Kulkarni, M.B.; Bhaiyya, M. Artificial intelligence and machine learning in infectious disease diagnostics: A comprehensive review of applications, challenges, and future directions. Microchem. J. 2025, 218, 115802. [Google Scholar] [CrossRef] [Scilit]
  20. Caliman Sturdza, O.A.; Filip, F.; Terteliu Baitan, M.; Dimian, M. Deep Learning Network Selection and Optimized Information Fusion for Enhanced COVID-19 Detection: A Literature Review. Diagnostics 2025, 15, 1830. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  21. Bilal, H.; Khan, M.N.; Khan, S.; Shafiq, M.; Fang, W.; Khan, R.U.; Rahman, M.U.; Li, X.; Lv, Q.-L.; Xu, B. The role of artificial intelligence and machine learning in predicting and combating antimicrobial resistance. Comput. Struct. Biotechnol. J. 2025, 27, 423–439. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  22. Orcales, F.; Tan, L.M.; Johnson-Hagler, M.; Suntay, J.M.; Ali, J.; Recto, K.; Glenn, P.; Pennings, P. Using genomic data and machine learning to predict antibiotic resistance: A tutorial paper. PLoS Comput. Biol. 2024, 20, e1012579. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  23. Zou, Z.; Tang, F.; Qiao, L.; Wang, S.; Zhang, H. Integrating sequencing methods with machine learning for antimicrobial susceptibility testing in pediatric infections: Current advances and future insights. Front. Microbiol. 2025, 16, 1528696. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  24. Talianu, A.; Fraser-Krauss, O.; Bolton, W.; Ming, D.; Zhu, N.; Hernandez, B.; Gilchrist, M.; Holmes, A.; Georgiou, P.; Rawson, T.M. Artificial intelligence for improving decision-making in bacterial infection management: A narrative review. J. Antimicrob. Chemother. 2026, 81, dkaf470. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  25. Singhal, N.; Vardhan, H.; Jain, R.; Gupta, P.; Pandey, A.; Wagri, N.K.; Gaur, A. Role of artificial intelligence in automating diagnostic procedures in clinical microbiology laboratories. Curr. Res. Biotechnol. 2025, 10, 100351. [Google Scholar] [CrossRef] [Scilit]
  26. Cesaro, A.; Hoffman, S.C.; Das, P.; de la Fuente-Nunez, C. Challenges and applications of artificial intelligence in infectious diseases and antimicrobial resistance. Antimicrob. Resist 2025, 3, 2. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  27. Antimicrobial Stewardship Interventions: A Practical Guide; Licence: CC BY-NC-SA 3.0 IGO; WHO Regional Office for Europe: Copenhagen, Denmark, 2021; Available online: https://www.who.int/europe/publications/i/item/9789289056267 (accessed on 7 April 2026).
  28. National Center for Emerging and Zoonotic Infectious Diseases (U.S.). Division of Healthcare Quality Promotion. The Core Elements of Hospital Antibiotic Stewardship Programs [2017]. Available online: https://stacks.cdc.gov/view/cdc/45312 (accessed on 7 April 2026).
  29. Gohil, S.K.; Septimus, E.; Kleinman, K.; Avery, T.R.; Varma, N.; Sands, K.E.; Mauricio, A.; Poland, R.E.; Guy, J.; Platt, R.; et al. Initial Antibiotic Selection Strategy and Subsequent Antibiotic Use-Insights from the INSPIRE Trials. JAMA 2025, 334, 1107–1109. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  30. Gohil, S.K.; Septimus, E.; Kleinman, K.; Varma, N.; Sands, K.E.; Avery, T.R.; Mauricio, A.; Sljivo, S.; Rahm, R.; Roemer, K.; et al. Improving Empiric Antibiotic Selection for Patients Hospitalized with Abdominal Infection: The INSPIRE 4 Cluster Randomized Clinical Trial. JAMA Surg. 2025, 160, 733–743. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  31. Gohil, S.K.; Septimus, E.; Kleinman, K.; Varma, N.; Avery, T.R.; Heim, L.; Rahm, R.; Cooper, W.S.; Cooper, M.; McLean, L.E.; et al. Stewardship Prompts to Improve Antibiotic Selection for Urinary Tract Infection: The INSPIRE Randomized Clinical Trial. JAMA 2024, 331, 2018–2028. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  32. Tao, L.; Zhang, C.; Zeng, L.; Zhu, S.; Li, N.; Li, W.; Zhang, H.; Zhao, Y.; Zhan, S.; Ji, H. Accuracy and Effects of Clinical Decision Support Systems Integrated with BMJ Best Practice–Aided Diagnosis: Interrupted Time Series Study. JMIR Med. Inf. 2020, 8, e16912. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  33. Touati, A.; Boufahja, F.; Ben Hamadi, N.; Touaitia, R.; Idres, T. Artificial Intelligence Applications in Antimicrobial Resistance: Comprehensive Review of Predictive Models, Diagnostic Innovations, and Clinical Integration. Microb. Drug Resist. 2026, 32, 422–443. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  34. Pinto, A.; Pennisi, F.; Ricciardi, G.E.; Signorelli, C.; Gianfredi, V. Evaluating the impact of artificial intelligence in antimicrobial stewardship: A comparative meta-analysis with traditional risk scoring systems. Infect. Dis. Now. 2025, 55, 105090. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  35. Bolton, W.J.; Wilson, R.; Gilchrist, M.; Georgiou, P.; Holmes, A.; Rawson, T.M. The impact of artificial intelligence-driven decision support on uncertain antimicrobial prescribing: A randomised, multimethod study. Lancet Digit. Health 2025, 7, 100912. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  36. Frenkel, A.; Rendon, A.; Chavez-Lencinas, C.; De la Torre, J.C.G.; MacDermott, J.; Gross, C.; Allman, S.; Lundblad, S.; Zavala, I.; Gross, D.; et al. Internal Validation of a Machine Learning-Based CDSS for Antimicrobial Stewardship. Life 2025, 15, 1123. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  37. Goodman, K.E.; Heil, E.L.; Claeys, K.C.; Banoub, M.; Bork, J.T. Real-world Antimicrobial Stewardship Experience in a Large Academic Medical Center: Using Statistical and Machine Learning Approaches to Identify Intervention “Hotspots” in an Antibiotic Audit and Feedback Program. Open Forum Infect. Dis. 2022, 9, ofac289. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  38. Fahim, Y.A.; Hasani, I.W.; Kabba, S.; Ragab, W.M. Artificial intelligence in healthcare and medicine: Clinical applications, therapeutic advances, and future perspectives. Eur. J. Med. Res. 2025, 30, 848. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  39. Lee, H.; Kim, Y.J.; Kim, J.H.; Kim, S.K.; Jeong, T.D. Optimizing Initial Vancomycin Dosing in Hospitalized Patients Using Machine Learning Approach for Enhanced Therapeutic Outcomes: Algorithm Development and Validation Study. J. Med. Internet Res. 2025, 27, e63983. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  40. Quigley, A.; Honeyman, D.; Stone, H.; Dawson, R.; MacIntyre, C.R. EPIWATCH, an artificial intelligence early-warning system as a valuable tool in outbreak surveillance. Int. J. Infect. Dis. 2025, 152, 107579. [Google Scholar] [CrossRef] [Scilit]
  41. Yang, J.; Eyre, D.W.; Lu, L.; Clifton, D.A. Interpretable machine learning-based decision support for prediction of antibiotic resistance for complicated urinary tract infections. Antimicrob. Resist 2023, 1, 14. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  42. Rani, S.; Kumar, R.; Panda, B.S.; Kumar, R.; Muften, N.F.; Abass, M.A.; Lozanović, J. Machine Learning-Powered Smart Healthcare Systems in the Era of Big Data: Applications, Diagnostic Insights, Challenges, and Ethical Implications. Diagnostics 2025, 15, 1914. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  43. Liu, G.Y.; Yu, D.; Fan, M.M.; Zhang, X.; Jin, Z.Y.; Tang, C.; Liu, X.F. Antimicrobial resistance crisis: Could artificial intelligence be the solution? Mil. Med. Res. 2024, 11, 7. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  44. Taylor, R.A.; Sangal, R.B.; Smith, M.E.; Haimovich, A.D.; Rodman, A.; Iscoe, M.S.; Pavuluri, S.K.; Rose, C.; Janke, A.T.; Wright, D.S.; et al. Leveraging artificial intelligence to reduce diagnostic errors in emergency medicine: Challenges, opportunities, and future directions. Acad. Emerg. Med. 2025, 32, 327–339. [Google Scholar] [CrossRef] [Scilit] [PubMed]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.