AI Models Detect Type 1 Diabetes Risk Before Onset

New Studies Highlight Machine Learning as Potential Early Risk Detection Strategy for Type 1 Diabetes

Each year, around 64,000 Americans are diagnosed with type 1 diabetes. Alarmingly, 40% remain unaware they have the disease until a life-threatening event occurs. This silent progression highlights the critical need for earlier detection methods.

AI Model Enhances Type 1 Diabetes Risk Assessment

Developments from two studies presented at the 85th Scientific Sessions of the American Diabetes Association (ADA) in Chicago demonstrate the potential of machine learning to identify type 1 diabetes risk up to a year before diagnosis. The models show significant improvements in accuracy and fewer false positives compared to standard screening methods.

Reducing False Positives and Increasing Accuracy

The first study developed two age-specific machine learning models—one for individuals aged 0–24 and another for those 25 and older—using medical claims and lab test data from NorstellaLinQ. The models identified confirmed cases of stage 3 type 1 diabetes with high sensitivity—80% in the younger group and 92% in adults. They also maintained improved precision, with a positive rate of just 0.3% in the general population.

Laura Wilson, director of health economics outcomes research at Sanofi, expressed enthusiasm about the study's implications: "By applying AI-driven predictive models to real-world data, we can help identify individuals at high risk much earlier, giving them the opportunity to plan and prepare for the future."

AI Detects Type 1 Diabetes Risk More Than 18-Fold

The second study used the Symphony Health Database, covering 75 million patients, to train a machine learning model. The model was tested on a large, real-world population and successfully identified people at risk for type 1 diabetes before symptoms appeared, increasing detection efficiency more than 18-fold. The best-performing model was Bidirectional Encoder Representations from Transformers (BERT), which correctly identified 80% of true type 1 diabetes cases.

Jared Joselyn, senior vice president and global head of E.D.G.E at Sanofi, emphasized the potential impact: "These findings show how AI can uncover hidden patterns in routine health care data and help improve detection rates, with the goal of fostering more proactive, scalable care before disease progression."

Next Steps and Future Research

Researchers plan to validate and refine a new clinical decision support tool for type 1 diabetes, working closely with leading hospital sites and experts. The research will integrate advanced AI models with hospital electronic health records, aiming to enable earlier, data-driven interventions for patients at risk.

Research Presentation Details

• Identification of Earlier Stage Autoimmune Type 1 Diabetes Using Machine Learning Algorithms**
• Presented on Sunday, June 22 at 12:30 p.m. CT

• Predictive Modeling for Presymptomatic Type 1 Diabetes Detection Using Open Claims Data**
• Presented on Sunday, June 22 at 12:30 p.m. CT

About the ADA's Scientific Sessions

The ADA's 85th Scientific Sessions, the world's largest scientific meeting focused on diabetes research, prevention, and care, will be held in Chicago, IL, on June 20–23. Thousands of leading physicians, scientists, and health care professionals are expected to attend, both in person and virtually, to unveil cutting-edge research and advances toward a cure for diabetes.

About the American Diabetes Association

The American Diabetes Association (ADA) is the nation's leading voluntary health organization fighting to end diabetes and helping people thrive. With 136 million Americans living with diabetes or prediabetes, the ADA drives discovery and research to prevent, manage, treat, and ultimately cure the disease. To learn more, visit diabetes.org or call 1-800-DIABETES (800-342-2383).