The AI system predicts patients at high risk of diabetic complications

More than 37 million people in the United States have diabetes, but many do not receive timely treatment, which can lead to costly and even fatal complications
.
While effective treatment is available in primary care settings, clinicians lack the necessary tools to identify those most
at risk.
To prevent adverse health outcomes before they occur, researchers at the University of Houston are developing primary care prediction, a clinical decision support system that uses deep learning to predict which patients are more likely to experience complications
.

The first tool developed in the innovative AI system is the Diabetes Complication Severity Index (DCSI) progression tool, which considers, in addition to a patient's medical history, how their social and environmental circumstances – employment status, living arrangements, education level, food security – increase their risk of
developing complications.
Studies have shown that these social factors influence the progression of
the disease.

Funded by the American Board of Family Medicine, the tool will provide clinicians with timely, actionable insights so they can intervene early, reduce the proportion of complications in people with diabetes, and reduce the number of
complications affecting each patient.

"Our long-term goal is to help clinicians be more proactive and less responsive
in treating diabetes.
By harnessing the power of artificial intelligence and machine learning, we can more effectively connect at-risk groups with interventions before their condition worsens," said
Winston Liaw, Ph.
D.
, principal investigator of the project and chair of the Department of Health Systems and Population Health Sciences at the Tilman J.
Fertitta School of Family Medicine.

For years, insurers and researchers have used DCSI to quantify complications
in patients at one point in time.
However, there are currently no tools to predict which people are most
at risk of rising DCSI scores.

The tool will be developed in collaboration with the Humana Institute for Integrative Health Systems Science at the University of Houston and leverage unique datasets from Humana — claims, health records, and social risk factors
for individuals and communities.
The tool will be tested within the PRIME Registry, a national platform
that includes millions of primary care patients across the country.

"The challenge with existing forecasting tools is that they provide little explanation and no guidance for follow-up, limiting trust and execution
.
The tools we are developing will tell clinicians why patients are at risk and recommend actions to reduce that risk," said
Ioannis Kakadiaris, Hugh Roy and Lillie Cranz Cullen University Professor of Computer Science, Health Systems and Population Health Sciences.

"Humana is excited to work with our University of Houston partners to leverage their expertise in artificial intelligence and predictive analytics, as well as our extensive diabetes experience
using DCSI and the Health Impact Social Decision Solution.
" This tool provides a great opportunity to put actionable information in the hands of primary care physicians, at the point of service, where real health change happens," said Dr.
Todd Prewitt, corporate medical director of Humana, clinical strategy and analysis
.

In addition to diabetes, the researchers believe the tool could also help predict complications associated with other diseases, such as uncontrollable high blood pressure or worsening depression
.
This tool will be particularly important
as the healthcare industry shifts to a value-based healthcare model.
In this model, doctors are rewarded for improving the health of their patients, rather than being paid
for every visit, surgery, or examination, regardless of the outcome.

Founded in 2019, Fertita School of Family Medicine has a social mission to improve health and healthcare in underserved urban and rural communities in Texas, with an emphasis on primary care education and research
.

"As primary care physicians, we need an effective way to use the wealth of information we receive to improve the quality of life
for our patients.
The number of complications experienced by patients is closely related to death or hospitalization, so developing such AI tools is crucial
.
”