Singapore General Hospital is developing an artificial intelligence solution that will enable it to determine the need for antibiotic prescriptions, limit their exploit and identify the most appropriate ones for each patient.
Developed using DXC technology, Augmented Intelligence in Infectious Diseases (AI2D) is an artificial intelligence model currently covering pneumonia cases. It was built on de-identified clinical data, including X-rays, clinical signs, vital signs and infection response trends, from approximately 8,000 SGH patients in 2019–2020. It included seven broad-spectrum intravenous antibiotics commonly prescribed for pneumonia.
CONCLUSIONS
Last year, a research team led by the SGH Pharmacy Division conducted a pilot validation study of the AI model, comparing it with 2,000 cases of pneumonia in 2023.
In their paper, SGH and DXC noted that AI2D helped reduce the number of cases for review threefold (624 from 2012). It also increased the ability to identify cases requiring intervention among this pile of audit cases by almost 12% compared to 4% with a full manual review process. What’s more, data analysis in one case that took 20 minutes with manual review was completed in “less than a second.”
In the study, the AI model was 90% right in determining the need for antibiotics in a given case of pneumonia; the study also found that almost 40% of antibiotics prescribed in these cases may have been unnecessary.
WHY IT’S IMPORTANT
SGH reported that pneumonia is the cause of 20% of all infections treated in their hospital, and of all infections, the most prescriptions for antibiotics are written. Patients stay in hospital on average for two to nine days, costing the government up to SGD5,000 (over $3,500) for each government-subsidized patient admitted to a public hospital.
Half of acute care hospitals worldwide are potentially prescribing inappropriate antibiotics to patients, resulting in antimicrobial resistance. At SGH, for example, the hospital unnecessarily prescribed 20-30% of broad-spectrum intravenous antibiotics, based on a 2018 antibiotic exploit audit. It is believed that in Singapore, as many as 30% of hospital infections have become resistant to broad-spectrum antibiotics.
Hospitals are establishing antimicrobial stewardship programs to address this growing global problem by preventing the overuse of antibiotics and identifying where they are best recommended – especially narrow-spectrum antibiotics. Such programs have the potential to support reduce hospital stays, minimize deaths and readmissions, and reduce patient and hospital costs.
In implementing antimicrobial stewardship programs, automation and artificial intelligence may be the best approach, as teams would require real-time visibility at the point of prescribing. Artificial intelligence would support identify cases requiring review and prioritize those requiring intervention. Automation would enable hospitals to expand the scope of auditing antibiotic exploit, providing right insight into antibiotic exploit.
The research team is currently preparing a comparative study involving 200 SGH hospital patients to check the effectiveness of the artificial intelligence model in reducing the exploit of antibiotics. Later, they will work to determine the most effective antibiotic against pneumonia, and then build the same model for urinary tract infection, another common hospital infection.
A BIGGER TREND
Outside Singapore, Taiwan’s China Medical University Hospital has been recognized for its efforts to combat antimicrobial resistance through the exploit of artificial intelligence. First rolled out in 2021 The Smart Antimicrobial System is a four-in-one AI-powered platform that identifies drug-resistant strains, predicts and monitors sepsis and death, recommends drug doses, and compares drug interactions. This specific exploit of artificial intelligence in the EMR system was highlighted during the hospital’s revalidation for Phase 7 of the HIMSS study A model for the adoption of electronic medical records in November last year.
ON RECORDING
“Well-intentioned physicians continually balance the risks and benefits of antibiotic exploit. It is often challenging to definitively determine whether patients will benefit from them based on clinical judgment, patient-specific factors, or the severity of the condition alone. If these antibiotics are not prescribed quickly in people who really need them, it can lead to sedate complications, but the misuse of antibiotics contributes to antibiotic resistance, which creates challenges for the future treatment of infections,” said Dr. Piotr Chlebicki, member of the AI2D project and senior consultant at the Department of Infectious Diseases of the Warsaw School of Economics. Diseases.