High accuracy diabetes prediction from an AI health care platform, but it “won’t replace patient care”
A Connecticut-based business seeks to lessen the burden of diabetes on both patients and physicians.
According to the Centers for Disease Control and Prevention, 37.3 million Americans have diabetes, 8.5 million of them are undiagnosed. Diabetes is one of the deadliest and most expensive illnesses in the nation. A new AI-based solution that seeks to lessen the burden of diabetes for both patients and doctors was unveiled on Tuesday by Greenwich, Connecticut-based Cedar Gate Technologies.
In a recent research, Cedar Gate used its product, Cedar Gate Analytics, to analyze the information from more than 1.2 million patients in its database over the course of a year.
According to the company’s findings, 80% of people who were identified by the model but had no history of diabetes were later found to have the condition.
Technology applications for business use cases
Cedar Gate Analytics, according to the business, is the “first commercially available and deployed value-based platform with this level of accuracy.”
In an interview with Fox News Digital, Cedar Gate’s chief product and business development officer Rajiv Mahale said that as a result, “we can confidently identify people who are at a higher risk of developing or being diagnosed with diabetes in the future.”
Although not the only AI-based model to identify diabetes risk, Cedar Gate Analytics claims to be “the first of its kind at this level.”
While there have been numerous university-based research studies demonstrating the effectiveness of AI-related diabetes prediction models, Mahale stated, “We’re applying our technology to commercial use cases, giving business users access to advanced analytics tools to solve real-world problems.”
“We don’t expect AI to replace patient care.”
He added, “Today, we have the technology to curb the trajectory of diabetes deeply and meaningfully, controlling cost and improving health outcomes at scale.”
The AI model developed by Cedar Gate is based on data from several years’ worth of medical claims. It makes predictions on the likelihood of several chronic illnesses, including diabetes, using machine learning networks.
This level of precision and predictability, albeit not novel, is thrilling, according to Snow.
According to leading industry analysts, there is currently no system that can achieve 80%+ accuracy at true, meaningful scale.
Before the data is entered into a model, “we have data scientists and clinical people creating meaning from the data, defining the patient’s journey and key events,” said Mahale. Our resources could help a much larger proportion of people who are at risk of developing significant chronic diseases.
Researchers believe that by identifying diabetes risk earlier and enhancing patient outcomes, the business might potentially contribute to the reduction of millions of dollars in medical costs.
According to the American Diabetes Association, the cost for each diabetic patient is $9,601 annually.
The application of artificial intelligence and machine learning in healthcare is not new, but it is expanding quickly as new avenues for patient care improvement and provider efficiency are made possible by the technology. The Food & Drug Administration (FDA) has a list of more than 500 AI/ML-enabled medical devices that are offered in the US as of October 2022.