Trinity Life Sciences, a leader in advisory, insights and analytics for the life sciences industry, presented results from a new study demonstrating the potential of an artificial intelligence and machine learning (AIML) model to detect type 1 diabetes (T1D) at least one year before a patient’s clinical diagnosis. Leveraging Trinity’s leading-edge AI capabilities and deep therapeutic expertise, with support from Sanofi, the model also maintains improved accuracy compared to traditional T1D screening methods.
The retrospective cohort study, Identification of Earlier Stage Autoimmune Type 1 Diabetes Using Machine Learning Algorithms, conducted by Trinity and Sanofi, aimed to develop a predictive machine learning model that could identify individuals at high risk of progression to stage 3 T1D. An integrated healthcare open claims (e.g., anonymized patient demographics, conditions, symptoms, treatments, etc.) and lab data set was used to develop the model. It was trained and validated using data from confirmed T1D patients, focusing on both pediatric and adult populations.
Also Read: Asepha Raises $4M in 5x Oversubscribed Seed Round for AI Pharmacy
Key findings of the study:
- With a 12-month blinded period built into its design, the model demonstrated the ability to identify potential T1D cases at least one year before formal diagnosis.
- The model balances sensitivity and accuracy, prioritizing the ability to capture as many true positive cases as possible while minimizing false positives.
- The model is more likely to identify adult T1D populations than pediatric ones, which differs from typical healthcare practices.
- By targeting higher risk individuals, the approach aims to modernize T1D screening strategies and improve patient outcomes.
“We are thrilled to collaborate with Sanofi on this groundbreaking study,” said Clare Gora, Partner & Head of Data Strategy & Analytics at Trinity Life Sciences. “At Trinity, we believe the intersection of AI technologies, life sciences expertise and data is where true innovation happens.”
Source: Businesswire
