Day Zero Diagnostics has introduced Keynome gAST, a genomic testing method that uses AI to speed up sepsis diagnosis by directly analyzing bacterial genomes from patient samples, potentially transforming treatment and reducing mortality. Researchers at Day Zero Diagnostics have developed Keynome gAST, an AI-based antimicrobial sensitivity genomic test that quickly predicts antimicrobial resistance by analyzing bacterial genomes from blood samples. This breakthrough, showcased at the ASM Microbe, could significantly enhance sepsis diagnosis and treatment, speeding up decision-making and potentially saving lives in the face of increasing antimicrobial resistance.

Sepsis is a potentially deadly infectious complication and accounts for 1.7 million hospitalizations and 350,000 deaths per year in the United States. A rapid and accurate diagnosis is crucial, as the risk of mortality increases by up to 8% every hour without effective treatment. However, the current diagnostic standard relies on culture growth, which typically takes 2 to 3 days. Doctors may choose to administer broad-spectrum antibiotics until more information is available for an accurate diagnosis, but these may have limited efficacy and potential toxicity for the patient.

Innovative AI Approach in Diagnosis

In a study presented at the ASM Microbe, a team from Day Zero Diagnostics unveiled a new approach to antimicrobial sensitivity testing using artificial intelligence (AI). Their system, Keynome gAST, or genomic antimicrobial sensitivity test, bypasses the need for culture growth by analyzing whole bacterial genomes extracted directly from patients’ blood samples. Preliminary results are based on studies that collected samples from 4 hospitals in the Boston area.

Revolutionizing Sepsis Treatment through Machine Learning

Unlike traditional methods that rely on known resistance genes, machine learning algorithms autonomously identify resistance and sensitivity factors based on data from a constantly expanding large-scale database of over 75,000 bacterial genomes and 800,000 sensitivity test results (48,000 bacterial genomes and 450,000 sensitivity test results at the time of this study). This allows for rapid and accurate predictions of antimicrobial resistance, revolutionizing sepsis diagnosis and treatment.

Future Directions and Implications

“The result is a first-of-its-kind demonstration of a comprehensive and high-precision prediction of antimicrobial sensitivity and resistance on clinical samples directly from blood,” said Jason Wittenbach, Ph.D., Director of Data Science at Day Zero Diagnostics and lead author of the study. “This represents a crucial demonstration of the feasibility of rapid antimicrobial resistance diagnosis based on machine learning, which could revolutionize treatment, reduce hospital stays, and save lives.” Researchers suggest that further studies are needed, given the limited sample size, but the results could contribute to significant advancements in patient outcomes in the face of the growing threat of antimicrobial resistance and the need for rapid sepsis diagnosis and treatment.

Funding for this research was partly provided by the Combating Antibiotic-Resistant Bacteria Biopharmaceutical Accelerator (CARB-X).

The development of AI-driven diagnostic tools like Keynome gAST holds immense promise in transforming the way sepsis is diagnosed and treated, offering hope for more effective and timely interventions that can ultimately save lives. This innovative approach represents a significant step forward in the fight against sepsis and antimicrobial resistance, highlighting the potential of AI in revolutionizing healthcare practices.