RenoTrue: A diabetes-specific machine learning model to estimate glomerular filtration rate for people with diabetes
Authors' conclusion
For people with diabetes, RenoTrue demonstrated better performance compared to the 2009 CKD-EPI equation in terms of estimating GFR across the full range of GFR.
Study context
- Population
- People with T1D and T2D from five international cohorts; median mGFR 75 ml/min per 1.73 m², median age 59 yearstest dataset 20%
- Participants
- 5,619 people
- Comparison
- RenoTrue vs 2009 CKD-EPI equation, against mGFR
- Outcome
- Performance vs mGFR in test dataset (agreement, bias, p30)
Methodological detailsmachine learning model development and validation, five international cohorts, training/validation/test split
- Machine learning model development and validation
- Five international cohorts
- Training/validation/test split
Performance
As reported in the abstract
| Test or model | ICC | Bias (ml/min per 1.73 m²) | p30 |
|---|---|---|---|
| RenoTrue | 0.87 (95% CI: 0.78, 0.93) | −0.57 (95% CI: −1.59, 0.46) | 81% (95% CI: 79%, 83%) |
| 2009 CKD-EPI | 0.86 (95% CI: 0.77, 0.92) | 4.17 (95% CI: 3.14, 5.20) | 74% (95% CI: 72%, 77%) |
ICC
RenoTrue0.87 (95% CI: 0.78, 0.93)
2009 CKD-EPI0.86 (95% CI: 0.77, 0.92)
Bias (ml/min per 1.73 m²)
RenoTrue−0.57 (95% CI: −1.59, 0.46)
2009 CKD-EPI4.17 (95% CI: 3.14, 5.20)
p30
RenoTrue81% (95% CI: 79%, 83%)
2009 CKD-EPI74% (95% CI: 72%, 77%)
Original publication
RenoTrue: A diabetes-specific machine learning model to estimate glomerular filtration rate for people with diabetes. Diabetes Res Clin Pract. Published 2 February 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →