Newswise — Antimicrobial resistance (AMR) has become one of the biggest public health crises of the 21st century, claiming an estimated 5 million lives each year and driving up healthcare costs around the world. Excessive use of antibiotics in human medicine, agriculture, and livestock production continues to accelerate the emergence of resistant bacteria, especially in low- and middle-income countries. Traditional diagnostic methods remain essential, but are often too slow and fragmented to keep pace with rapidly evolving pathogens. Meanwhile, health systems are finding it increasingly difficult to integrate vast amounts of genomic, clinical, and epidemiological data. In the face of these mounting challenges, researchers are exploring AI-driven tools to predict resistance patterns, optimize antibiotic use, and enhance early detection and intervention strategies.
A research team from Peking Union Medical College Hospital and Xiangya Third Hospital of Central South University conducted a comprehensive review (DOI: 10.12290/xhyxzz.2025-0655). Medical journal of Peking Union Medical College Hospital (September 2025) reveals how AI is revolutionizing AMR prevention and control. This article describes how machine learning and deep learning are transforming surveillance, diagnosis, treatment optimization, and drug discovery, and provides a timely blueprint for integrating intelligent systems into global infection control.
This review details how AI technologies are being applied in four key areas of AMR prevention. First, in epidemiological surveillance and early warning, AI algorithms such as XGBoost analyze hospital resistance records and antibiotic consumption data to predict future outbreaks and help health authorities act before the crisis escalates. Natural language processing systems can also scan electronic records and social media to detect resistance “hot spots” in real time. Second, for resistance detection and prediction, AI-powered models trained on MALDI-TOF mass spectrometry and genomic data can identify resistant bacteria within hours, much faster than traditional culture testing. The model, trained on more than 300,000 bacterial samples, achieved high predictive accuracy. Staphylococcus aureus and Klebsiella pneumoniaeindicating clinical readiness. Third, in clinical decision-making, AI-based systems can reduce mismatched antibiotic prescriptions by up to half, promoting rational drug use in hospitals. Finally, in drug discovery, deep learning models, such as those that identified halicin and abausin, are uncovering entirely new classes of antibiotics with unique mechanisms. These AI advances are redefining how humanity detects, treats, and prevents resistance globally.
“AI is transforming the fight against antimicrobial resistance from reactive to predictive,” said corresponding author Dr. Li Zhang. “By integrating genomic, clinical, and environmental data, AI systems can uncover hidden infection patterns and recommend personalized treatments faster than ever before. But to achieve maximum impact, we also need to improve data quality, ensure algorithm transparency, and strengthen ethical oversight. Through cross-disciplinary collaboration, AI can bridge the gap between innovation and implementation, turning smart technologies into life-saving public health tools.”
The convergence of AI and infectious disease science signals a paradigm shift in global health defense. In hospitals, AI-powered diagnostic and decision support tools can help clinicians provide faster, more targeted treatment, reduce antibiotic misuse, and improve patient outcomes. On a broader scale, predictive analytics can guide surveillance and resource allocation and facilitate early containment of resistant pathogens. In pharmaceutical research, AI accelerates drug discovery by exploring the chemical space beyond human intuition. As technology continues to evolve, standardizing data, building interpretable models, and fostering global collaboration will be essential. AI is poised to become the foundation for precision infection control and sustainable healthcare.
###
References
Toi
10.12290/xhyxzz.2025-0655
Original source URL
https://xhyxzz.pumch.cn/article/doi/10.12290/xhyxzz.2025-0655
Funding information
National Natural Science Foundation of China (No. 82472341); Public Competition Program of the Chinese Medical Foundation (No. 23-520); Medical and Health Science Technology Innovation Engineering Project of the Chinese Academy of Medical Sciences (No. 2021-I2M-1-044).
About Medical journal of Peking Union Medical College Hospital
Medical journal of Peking Union Medical College Hospital is a leading clinical medical publication supported by the multidisciplinary expertise of Peking Union Medical College Hospital. Featuring the latest research, advances, and academic trends in clinical medicine, translational medicine, pharmacy, and related interdisciplinary fields, it is aimed at clinicians and medical students across China. This journal aims to foster the exchange of medical knowledge, lead academic discussions, and serve as a high-quality platform for promoting scholarly debate in clinical medicine. This journal is listed in China Science and Technology Core Journals (CSTPCD), China Science Citation Database (CSCD), China Core Journal Guide, and China Biomedical Literature Database (CMCC). Full-text content can be accessed on platforms such as Wanfang Data, CNKI, and Chongqing VIP Database. Indexed in Scopus (Netherlands), the Swedish Directory of Open Access Journals (DOAJ), and the Japan Science and Technology Agency (JST).
