Singapore General Hospital has implemented a predictive AI tool to help check the health status of patients for surgery.
Dubbed CARES-ML (Combined Risk Assessment in Surgery-Machine Learning), the predictor generates surgical risk reports and scores using a patient’s medical history, analysis type of physical condition and investigational test results, including X-rays and blood tests. It achieved more than 90% accuracy in predicting surgical risks during hospital trials.
The tool is based on the CARES calculator that SGH developed in 2018. Since then, the tool has been refined and validated on a local dataset of nearly 100,000 surgical patients between 2015 and 2022. .
WHY IT IMPORTANT
According to SGH, CARES-ML can be used to improve preoperative evaluation. CARES-ML assists anesthesiologists and surgeons in assessing individual patients and enhancing clinical team decision-making and making recommendations on perioperative care plans for patients. This will ultimately improve patient outcomes and enhance patient safety,” added Hairil Rizal, principal investigator and associate professor and senior advisor of Anesthesiology at SGH.
The SGH team is currently working on expanding the AI model to predict a patient’s length of hospital stay as well as a patient’s risk of pneumonia and stroke. They’re also looking to leverage general AI like ChatGPT to assign a patient’s physical condition classification, which CARES-ML is also currently doing.
TREND TO BIGGER WOMAN
To ensure patients are fit for surgery, the Adelaide Center Local Health Network in Australia has developed a pre-operative rehabilitation program called My PreHab, which includes a comprehensive wellness questionnaire and a personalized checklist of items. It is being distributed through digital avenues designed by Personify Care.
AI is also being leveraged to monitor the condition of patients after surgery. Recently, Manipal hospitals in India have partnered with Singaporean startup ConnectedLife to develop a virtual platform in conjunction with Fitbit for continuous remote patient monitoring after surgery.
