.png){kind=logo}
Use clinical variables from the EMR of patients to predict antibiotic resistance.
Antimicrobial resistance (AMR) has become a serious global health problem. In intensive care units, the severity of the threat of ADR is highlighted by the fact that between 30% and 60% of antibiotics prescribed are considered unnecessary, inappropriate or suboptimal (1)(2). This high level of misuse not only undermines the effectiveness of treatments but also catalyzes the advancement of resistant bacterial strains. This is a significant and growing problem in Latin America, with increasing rates of resistance to key antibiotics [3].
For example, the non-susceptibility of Klebsiella pneumoniae to carbapenem antibiotics has been rising, reaching an average of 21% in recent years. Methicillin-resistant Staphylococcus aureus (MRSA) is prevalent in Latin America, with some studies indicating a prevalence as high as 48.3% in certain regions [4,7].
AI can process vast amounts of data from various sources, including electronic health records and laboratory results, to identify patterns and trends in antimicrobial resistance. This capability allows for real-time surveillance and early detection of emerging resistance threats
To address the issue of antimicrobial resistance, the “AMRForecast AI” model has been devised to guide the medical community in prescribing precise antibiotic therapies. By analyzing clinical and genomic data, AMRForecast AI makes it possible to predict the effectiveness of antibiotics.
The model integrates two main data sets. Genomic Data Resource (9) collects genomic sequences from clinical laboratories, essential for understanding pathogen evolution. Unstructured clinical records come from patients' EHRs (7) and contain a wealth of information including demographics, diagnoses, and treatment outcomes. These data sets provide a comprehensive view of antimicrobial resistance, allowing AI to detect and analyze resistance trends.
Antimicrobial resistance (AMR) has become a serious global health problem. In intensive care units, the severity of the threat of ADR is highlighted by the fact that between 30% and 60% of antibiotics prescribed are considered unnecessary, inappropriate or suboptimal (1)(2). This high level of misuse not only undermines the effectiveness of treatments but also catalyzes the advancement of resistant bacterial strains. This is a significant and growing problem in Latin America, with increasing rates of resistance to key antibiotics [3].
For example, the non-susceptibility of Klebsiella pneumoniae to carbapenem antibiotics has been rising, reaching an average of 21% in recent years. Methicillin-resistant Staphylococcus aureus (MRSA) is prevalent in Latin America, with some studies indicating a prevalence as high as 48.3% in certain regions [4,7].
AI can process vast amounts of data from various sources, including electronic health records and laboratory results, to identify patterns and trends in antimicrobial resistance. This capability allows for real-time surveillance and early detection of emerging resistance threats
To address the issue of antimicrobial resistance, the “AMRForecast AI” model has been devised to guide the medical community in prescribing precise antibiotic therapies. By analyzing clinical and genomic data, AMRForecast AI makes it possible to predict the effectiveness of antibiotics.
Economic evaluations of antimicrobial stewardship programs, which often incorporate AI tools, have shown inpatient net savings ranging from $540 to $24,231 for each prevented death. These programs optimize antibiotic use, thereby reducing the incidence of resistant infections and associated treatment costs [8].
AI can assist in selecting the most effective antibiotic treatment for individual patients by analyzing their specific infection profiles and resistance patterns. This personalized approach reduces the misuse of antibiotics, a key factor in the development of AMR.
The model integrates two main data sets. Genomic Data Resource (9) collects genomic sequences from clinical laboratories, essential for understanding pathogen evolution. Unstructured clinical records come from patients' EHRs (7) and contain a wealth of information including demographics, diagnoses, and treatment outcomes. These data sets provide a comprehensive view of antimicrobial resistance, allowing AI to detect and analyze resistance trends.
