Multimodal AI correlates of glucose spikes in people with normal glucose regulation, pre-diabetes and type 2 diabetes
Study context
- Population
- Prospective cohort of 1,137 participants in the United States; multimodal data analysed from deeply phenotyped individualsdeeply phenotyped
- Participants
- 347 participants
- Exposure
- Normoglycemic vs prediabetic vs T2D
- Normoglycemicn = 174reference
- Prediabeticn = 79
- T2Dn = 94
- Outcome
- Glucose spike metrics by diabetes state
Methodological detailsmultimodal data, deeply phenotyped, independent validation cohort
- Multimodal data
- Deeply phenotyped
- Independent validation cohort
Main finding
As reported in the abstract
- Significant differences in distribution of glucose spike metrics among diabetes states
- Longer expected time for spike resolution and higher nocturnal hypoglycemia values in T2D
Observational data: associations, not effects.
Further findings
Correlations of glycemic metrics with other modalities
- Significant correlation between mean glucose level and gut microbiome diversity
- Significant correlation between expected time for spike resolution and resting heart rate
Multimodal glycemic risk profiles, external validation
- Improved risk stratification, showing substantial variability among individuals with the same HbA1c
- Validated in 1,955 normoglycemic and 114 prediabetic individuals from an independent cohort
Original publication
Multimodal AI correlates of glucose spikes in people with normal glucose regulation, pre-diabetes and type 2 diabetes. Nat Med. Published 31 July 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →