Predicting Postprandial Glycemic Responses With Limited Data in Type 1 and Type 2 Diabetes
Authors' conclusion
We demonstrate that in individuals with T1D and T2D, glycemic responses to meals can be predicted without personalized training data or invasive physiological data.
Study context
- Population
- People with T1D or T2D who wore continuous glucose monitors (CGMs) and logged meals
- Participants
- 397 vs 100
- Comparison
- T1DEXI vs ShanghaiT2DM
- T1DEXIn = 397
- ShanghaiT2DMn = 100
- Primary endpoint
- Prediction of 2-hour PPGR and peak 2-hour postprandial glucose rise (Glumax) from dietary, demographic and temporal features
- Follow-up
- 2 hours
Methodological detailssecondary analysis of 2 datasets (T1DEXI, ShanghaiT2DM)
- Secondary analysis of 2 datasets (T1DEXI, ShanghaiT2DM)
Primary result
As reported in the abstract
Prediction accuracy (R) · 2 hours postprandial
| Group | PPGR R | Glumax R |
|---|---|---|
| T1DEXI | 0.61 | 0.64 |
| ShanghaiT2DM | 0.72 | 0.73 |
PPGR R
T1DEXI0.61
ShanghaiT2DM0.72
Glumax R
T1DEXI0.64
ShanghaiT2DM0.73
Further findings
| Outcome | P value |
|---|---|
| Sources of intraindividual variability (duplicate meals): time of day (PPGR)T1DEXI | P < 0.05 for lunch |
| Sources of intraindividual variability (duplicate meals): time of day (PPGR)ShanghaiT2DM | P < 0.001 for lunch and dinner |
| Sources of intraindividual variability (duplicate meals): menstrual cycle (Glumax) | P < 0.05 for perimenstrual |
Food category features
- Including food category features led to higher accuracy than macronutrients alone
Original publication
Predicting Postprandial Glycemic Responses With Limited Data in Type 1 and Type 2 Diabetes. J Diabetes Sci Technol. Published 5 March 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →