|
Application of Artificial Intelligence in Predicting Antibiotic Resistance in Clinical Samples: A Narrative Review
|
Ghazaal Ataei    , Ehsan Shirzad , Mahnaz Mohtashami * |
| Neyshabur Branch, Islamic Azad University , mahnazmohtashami@iau-neyshabur.ac.ir |
|
|
Abstract: (4 Views) |
Background and Objective: Antibiotic resistance is one of the major global public health challenges, contributing to treatment failure, prolonged hospitalization, increased healthcare costs, and mortality. In recent years, artificial intelligence (AI), particularly machine learning algorithms, has emerged as a novel approach for the identification and prediction of antibiotic resistance and for supporting clinical decision-making. This study aimed to review the available evidence on the applications of AI in the identification, prediction, and surveillance of antibiotic resistance.
Methods: This study was conducted as a narrative review. A literature search was performed in the PubMed, Scopus, and Google Scholar databases using the keywords Artificial Intelligence, Antibiotic Resistance, and Bacterial Resistance. Articles published between 2014 and 2026 were reviewed, and studies addressing the application of AI in the identification, diagnosis, prediction, and surveillance of antibiotic resistance were included.
Results: The reviewed studies demonstrated that various AI-based methods and tools, including machine learning, deep learning, convolutional neural networks, and graph neural networks, have been applied to the identification of antibiotic resistance genes and markers, prediction of resistance patterns, rapid detection of resistance, and surveillance of antibiotic resistance trends. These approaches have utilized genomic, metagenomic, laboratory, molecular, and clinical data to identify resistance-associated patterns and support decision-making in the management of infections caused by resistant pathogens. However, data heterogeneity, differences in laboratory methods, limited availability of standardized datasets, low interpretability of some models, and a lack of external validation studies remain among the major reported challenges.
Conclusion: Artificial intelligence has considerable potential to support the identification, prediction, and surveillance of antibiotic resistance and to facilitate therapeutic decision-making. However, it currently cannot replace standard laboratory methods. The development of standardized databases, multicenter studies, and external validation of AI models is essential to enhance the clinical applicability of these technologies. |
|
| Keywords: Artificial intelligence, Machine learning, Antibiotic resistance, Bacterial resistance, Clinical microbiology. |
|
|
|
Type of Study: Review |
Subject:
medicine, paraclinic
Received: 2026/06/12 | Revised: 2026/10/13 | Accepted: 2026/10/13
|
|
|
|
|
|
|
| Send email to the article author |
|
|
|
| Add your comments about this article |
|
|
|
_-_Copy.png)
.png){kind=logo}
