Getting personalized antibiotic prescriptions with just a few mouse clicks

Singapore, 18 November 2024 – Doctors may soon use an AI-driven solution to swiftly prescribe a personalised antibiotic regimen for patients with just a few mouse clicks instead of giving general treatment. The antibiotic regimen can then be adjusted, if necessary, when bacterial culture and other investigation test results becomes available.

Known as the Augmented Intelligence in Infectious Diseases (AI²D), the solution is developed by Singapore General Hospital (SGH) in collaboration with DXC Technology and Synapxe. It targets two common infections treated in hospitals - pneumonia and urinary tract infection. The team, led by SGH’s Division of Pharmacy, has developed the model for pneumonia which showed 90 per cent accuracy in a pilot validation study to firstly determine if antibiotics were necessary. They are now looking to measure its effectiveness in safely reducing antibiotic use in a comparative study involving 200 inpatients. These patients will be randomly assigned to have either doctors or doctors with AI²D determine whether antibiotics should be given.

“Well-intended doctors are constantly balancing the risks and benefits of antibiotic use. It is often hard to tell definitively that patients will benefit from it based on clinical assessment, patient-specific factors, or condition severity alone. If not prescribed promptly for those who genuinely needs them, it may lead to dire complications. Yet, misusing antibiotics contributes to antibiotic resistance, posing challenges for future infection treatment. So, a tool like AI²D will be very helpful to guide doctor’s decision before lab results are available,” said Dr Piotr Chlebicki, Senior Consultant, Department of Infectious Diseases, SGH and project member.

The pneumonia model was built using retrospective deidentified clinical data like x-rays, clinical symptoms, periodic vital signs, and trends of common body responses to infection, of about 8,000 SGH patients between 2019 and 2020. It was then validated against another 2,000 cases in 2023 using a design that would simulate real-life usage when deployed. Besides showing a high level of accuracy, the pilot validation study also revealed that almost 40 per cent of antibiotics prescribed in those cases to treat pneumonia at the onset may not have been necessary. This situation is not unique to Singapore. The Centers for Disease Control and Prevention (CDC) in U.S. estimated that up to 50 per cent of all antibiotics prescribed in the U.S. are unnecessary or inappropriate, with many of them prescribed in inpatient settings.

The team will then work on determining which type of antibiotic is most effective against pneumonia and the urinary tract infection model after the comparative study.

“AI²D can help inform doctors if antibiotics are needed, and if so, to recommend what is needed and the dosing regimen based on the patient’s relevant clinical data. This may potentially cut antibiotic use, which in turn lowers the risk of antibiotic resistance, keeping antibiotics effective for treating infections now and in the future with the help of AI,” said Associate Professor Andrea Kwa, Deputy Director, Pharmacy (Research and Innovation), SGH, and project lead.