Getting your Trinity Audio player ready...
|
In a significant leap forward, Singapore General Hospital (SGH) has embraced the power of artificial intelligence (AI) with the implementation of CARES-ML (Combined Assessment of Risk Encountered in Surgery – Machine Learning), an intelligent calculator designed to enhance accuracy in pre-surgery risk assessments.
CARES-ML, an evolved version of the 2017 AI tool CARES, utilises machine learning (ML) to process and interpret data from 100,000 SGH patients. The tool amalgamates pre-surgical assessments, calculating patient risks and generating a comprehensive report, including a risk score. Higher scores indicate an elevated risk of negative outcomes post-surgery.
Associate Professor Hairil Rizal, Senior Consultant in the Department of Anaesthesiology at SGH, highlighted the tool’s capabilities, stating, “We have a 90% certainty that the AI engine will be able to tell whether the patient needs ICU.”
Additionally, CARES-ML boasts an 83–86% accuracy in predicting 30-day mortality post-surgery. Especially, CARES-ML operates as a human-in-the-loop system, providing decision support while allowing clinicians to make the final risk assessment based on their professional judgment.
The continuous learning aspect of the tool ensures improved accuracy over time. As new data is incorporated, CARES-ML evolves, showcasing the adaptability and efficacy typical of advanced AI engines.
This AI-driven innovation addresses the inherent risks associated with surgeries, as 17% of more than 300 million major surgeries worldwide annually, pre-COVID-19, experienced complications. SGH’s proactive approach involves pre-surgery assessments conducted approximately 10 days before the scheduled procedure. For low-risk patients, surgeries typically proceed, while interventions are initiated for those in higher-risk categories.
CARES-ML aids in identifying modifiable risk factors and optimising patient conditions, offering the medical team valuable insights. By pinpointing factors like anaemia early on, patients can receive timely treatment. The tool plays a crucial role in decision-making, especially for higher-risk patients, allowing for considerations such as less invasive alternatives, nonsurgical options, or potential surgery postponements.
One of the key advantages of AI in this context is its ability to analyse vast amounts of data rapidly. Just before surgery, the latest health data is input into CARES-ML, which, combined with doctors’ expertise, fine-tunes the risk score. The AI’s capability to predict the need for ICU resources is particularly noteworthy, enabling better resource management and optimal patient outcomes.
SGH plans to expand CARES-ML’s capabilities by developing estimates for other complications, including heart attacks, respiratory issues, and kidney failure. To substantiate the impact of CARES-ML on healthcare outcomes, SGH will conduct the IMAGINATIVE Trial, a randomised control trial involving 9,000 patients.
Funded by the National Medical Research Council, the trial aims to assess the impact of CARES-ML deployment on complication rates, mortality, and ICU utilisation, with results expected by 2026. The implementation of AI in surgical risk assessment not only exemplifies technological advancement but also holds the promise of transforming patient outcomes and healthcare delivery.
Earlier, OpenGov Asia reported that SGH along with the Nanyang Technological University, Singapore (NTU Singapore) is set to establish a Joint Research & Development Laboratory in additive manufacturing (3D printing). This partnership aims to usher in a new era of healthcare solutions, particularly personalised medical devices and implants, benefitting SGH patients.
The collaboration capitalises on the synergies between SGH’s 3D Printing Centre and NTU’s Singapore Centre for 3D Printing (SC3DP). The envisioned innovations are still in the developmental phase but hold promise for redefining patient treatment.