AI Breakthrough: Boosting Precision in UTI Antibiotic Prescribing to Combat Antimicrobial Resistance

New research by scientists at the University of Liverpool investigates how artificial intelligence (AI) can assist doctors in antibiotic prescription for urinary tract infections (UTIs), a common bacterial infection. The research aims to enhance the precision of antibiotic prescribing to address the global crisis of antimicrobial resistance (AMR).

AI-Powered Precision Prescribing

This work presents an AI-based method for treatment decision-making that integrates human judgment with data-driven predictions. The algorithm uses a mathematical tool, known as a utility function, to evaluate the benefits and drawbacks of various antibiotic options for individual patients.

The system seeks to decrease the unnecessary use of strong antibiotics and slow the development of resistance by selecting appropriate antibiotics.

The Global Threat of Antimicrobial Resistance (AMR)

Dr. Alexander Howard from the University of Liverpool's Department of Pharmacology & Therapeutics stated that AMR is a significant global public health threat.

Bacterial AMR was directly responsible for an estimated 1.27 million global deaths in 2019 and contributed to 4.95 million deaths.

He noted the necessity for innovative solutions to facilitate the precise use of antimicrobials.

Promising Findings and Built-in Safety

A simulation study, which utilized real healthcare data, found that the AI's recommendations were comparable to those made by human doctors.

The AI's suggestions were less likely to contribute to antibiotic resistance and more frequently recommended oral antibiotics over intravenous ones. The algorithm also includes a safety feature that prioritizes an effective antibiotic selection when a patient is critically ill.

Future Outlook and University Contributions

Further research is required across diverse global settings to validate the wider applicability of these findings, particularly in regions most affected by antibiotic resistance. This work contributes to the University of Liverpool's research in Therapeutics Innovation and Infection Resilience.

The paper, titled 'Algorithmic antibiotic decision-making in urinary tract infection using prescriber-informed prediction of treatment utility,' was published in npj Digital Medicine.