Skip to main content

Predictive AI in Clinical Practice

Today, the ability to predict complications of type 2 diabetes is supported by published clinical evidence.

The EGOAL-AMD study, published in Diabetes Research and Clinical Practice, evaluated a machine learning algorithm integrated into the MetaClinic® electronic medical record.

The algorithm estimates the 2-year or 5-year risk of six chronic complications:

    • Cardiovascular disease
    • Cerebrovascular disease
    • Peripheral vascular disease
    • Nephropathy
    • Neuropathy
    • Eye complications

Risk levels are displayed on a color-coded scale directly during the consultation.
The study involved 138,558 patients across 38 Italian diabetes centers between April 2023 and December 2024. Among the approximately 20,000 patients for whom physicians requested a predictive risk assessment, the findings showed a reduction in therapeutic inertia: making risk visible encouraged.