Reviewed by Lexie CornerJun 17 2024
In a study presented at ASM Microbe, Day Zero Diagnostics researchers revealed a revolutionary technique for antimicrobial susceptibility testing based on artificial intelligence (AI). Their approach, the Keynome® gAST, or genomic Antimicrobial Susceptibility Test, analyzes bacterial whole genomes collected directly from patient blood samples, eliminating the requirement for culture growth.
Sepsis is a life-threatening infection complication that leads to 1.7 million hospitalizations and 350,000 fatalities in the United States every year. Fast and precise diagnosis is crucial since the risk of death increases by up to 8 % every hour without proper treatment. However, the current diagnostic standard is based on culture growth, which usually takes 2–3 days.
Doctors can provide broad-spectrum antibiotics until further information is available for a correct diagnosis, but they have limited effectiveness and could harm the patient.
The interim findings are based on data obtained from four Boston-area hospitals.
As opposed to conventional techniques that depend on known resistance genes, the machine learning algorithms independently determine the drivers of susceptibility and resistance using information from an ever-expanding large-scale database that contains over 800,000 susceptibility test results and over 75,000 bacterial genomes (48,000 bacterial genomes and 450,000 susceptibility test results at the time of this study).
This transforms the diagnosis and management of sepsis by enabling quick and precise predictions of antimicrobial resistance.
The result is a first-of-its-kind demonstration of comprehensive and high-accuracy antimicrobial susceptibility and resistance predictions on direct-from-blood clinical samples. This represents a critical demonstration of the feasibility of rapid machine learning-based diagnostics for antimicrobial resistance that could revolutionize treatment, reduce hospital stays, and save lives.
Jason Wittenbach, Ph.D., Study Lead Author and Director, Data Science, Day Zero Diagnostics
Given the small sample size, the researchers believe further research is needed. However, the findings presented could lead to considerable improvements in patient outcomes, given the growing issue of antimicrobial resistance and the necessity for fast detection and treatment of sepsis.
This research was partially funded by the Combating Antibiotic-Resistant Bacteria Biopharmaceutical Accelerator (CARB-X).