Application of Artificial Intelligence in Predicting Bacterial Antibiotic Resistance with a Focus on Deep Learning Models and the Host Immune System
محورهای موضوعی : Biotechnological Journal of Environmental MicroorganismsElisa Farahmand Jahankhanmlu 1 , Hoora Moharrami Nasrabadi 2
1 - Department of Microbiology, Ard.C., Islamic Azad University, Ardabil, Iran
2 - Department of Microbiology, Ard.C., Islamic Azad University, Ardabil, Iran
کلید واژه: Antimicrobial Resistance (AMR), Artificial Intelligence (AI), Host Immune System, Clinical Decision Support Systems, Predictive Modeling, Bacterial Pathogens,
چکیده مقاله :
Antibiotic resistance is a significant global health issue. It is driven by the fast evolution of bacterial pathogens and the complex interactions between microbes and the host immune system. Artificial intelligence, especially deep learning, has become an important tool for analyzing large biological and clinical datasets. It can predict antibiotic resistance patterns more accurately and quickly. This review summarizes recent advancements in AI-driven resistance prediction, focusing on deep learning models and their ability to incorporate host immune responses as predictive markers. The review gathers findings from recent studies published over the last decade. It includes research on machine learning and deep learning methods for predicting antimicrobial resistance (AMR). The sources cover genomic and phenotypic bacterial data, clinical information, electronic health records, immunological markers, and multi-omics datasets. Particular attention is given to models that consider the dynamics of the host immune system. Evidence shows that deep learning architectures, such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), graph neural networks (GNNs), and transformer-based models, achieve high accuracy in predicting AMR from genomic sequences, bacterial gene expression, and clinical variables. Including host immune signatures, such as cytokine profiles, inflammatory markers, and immune cell activity, further improves predictive performance. This is especially true when distinguishing between resistant and susceptible infections. Multimodal AI models that use both microbial and host features tend to generalize better than single-source models. However, challenges remain. These include data variety, limited access to curated immune datasets, and the need for clear AI models that can be used in clinical settings. Deep learning-based AI provides a new way to predict bacterial antibiotic resistance, especially when combined with information about the host immune system. These methods could enhance diagnostic accuracy, tailor antimicrobial therapy, and aid real-time clinical decision-making. Progress in this field will require larger multimodal datasets, greater transparency of models, and thorough clinical validation to integrate AI-driven AMR prediction into everyday medical practice.
Antibiotic resistance is a significant global health issue. It is driven by the fast evolution of bacterial pathogens and the complex interactions between microbes and the host immune system. Artificial intelligence, especially deep learning, has become an important tool for analyzing large biological and clinical datasets. It can predict antibiotic resistance patterns more accurately and quickly. This review summarizes recent advancements in AI-driven resistance prediction, focusing on deep learning models and their ability to incorporate host immune responses as predictive markers. The review gathers findings from recent studies published over the last decade. It includes research on machine learning and deep learning methods for predicting antimicrobial resistance (AMR). The sources cover genomic and phenotypic bacterial data, clinical information, electronic health records, immunological markers, and multi-omics datasets. Particular attention is given to models that consider the dynamics of the host immune system. Evidence shows that deep learning architectures, such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), graph neural networks (GNNs), and transformer-based models, achieve high accuracy in predicting AMR from genomic sequences, bacterial gene expression, and clinical variables. Including host immune signatures, such as cytokine profiles, inflammatory markers, and immune cell activity, further improves predictive performance. This is especially true when distinguishing between resistant and susceptible infections. Multimodal AI models that use both microbial and host features tend to generalize better than single-source models. However, challenges remain. These include data variety, limited access to curated immune datasets, and the need for clear AI models that can be used in clinical settings. Deep learning-based AI provides a new way to predict bacterial antibiotic resistance, especially when combined with information about the host immune system. These methods could enhance diagnostic accuracy, tailor antimicrobial therapy, and aid real-time clinical decision-making. Progress in this field will require larger multimodal datasets, greater transparency of models, and thorough clinical validation to integrate AI-driven AMR prediction into everyday medical practice.
1) Nayak, D. S., Pati, A., Panigrahi, A., Moharana, B., Rath, A., Sahoo, S. K., & Swarnkar, T. (2023). Artificial intelligence in battle against antimicrobial resistance: opportunities and challenges. Int J Recent Innov Trends Comput Commun, 11, 427-437.
2) Sakagianni, A., Koufopoulou, C., Koufopoulos, P., Feretzakis, G., Anastasiou, A., Theodorakis, N., & Myrianthefs, P. (2025). Influence of Microbiome Interactions on Antibiotic Resistance Development in the ICU Environment: Insights and Opportunities with Machine Learning. Acta Microbiologica Hellenica, 70(2), 14.
3) Abhijit, G., & Mishra, V. (2024). Artificial Intelligence Tools to Address Challenges of Antimicrobial Resistance in Pathogenic Biofilm Systems. Available at SSRN 4970499.
4) Sun, S. (2025). Progress in the Identification and Design of Novel Antimicrobial Peptides Against Pathogenic Microorganisms. Probiotics and Antimicrobial Proteins, 17(2), 918-936.
5) Wang, X. W., Wang, T., & Liu, Y. Y. (2024). Artificial Intelligence for Microbiology and Microbiome Research. arXiv preprint arXiv:2411.01098.
