964-P: Reducing Discordance between GMI and A1C Using AI
Bibliographic record
Abstract
Introduction and Objective: GMI is increasingly relied upon in clinical practice as a surrogate for A1c. However absolute discordance >0.5% is common between GMI & A1c. GMI also tends to overestimate at lower A1c (<7.0) and underestimate at higher A1c (>8.0). BCDiabetes, a public Canadian clinic, has permissioned access to anonymized client data: more than 3,000 of its clients use CGM 365 days per year & have A1c measured intermittently. Methods: Prior to 2024-Sep-30, 2922 pairs of 90 days of CGM data & A1c values measured on the 90th day of CGM from 1508 patients, were used to build two machine learning (ML) models designed to minimize discordance in GMI vs A1c. Post 2024-Sep-30, 270 analogous pairs from 268 different clients with 90 days of CGM data & A1c values, were used to test the two ML models against the Bergenstal equation (2018) & an analogous linear regression equation developed with the same training data. Results: The Bergenstal equation showed a mean absolute difference of 0.43, with 37% of participants demonstrating discordance; the BCDiabetes equation was minimally higher than Bergenstal. Both ML models outperformed the Bergenstal equation with ML Model 2 reducing the relative risk of discordance compared to Bergenstal by 18% (Figure 1). Conclusion: AI modeling of GMI vs A1c promises to reduce discordance. With validation of such models and their adoption into clinical practice, GMI will become a more reliable indicator of A1c. Disclosure A. Alqahtani: None. G. Netto Flores Cruz: Employee; BiomeHub. N. Khan: None. K. Hawke: None. T. Elliott: None.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.010 | 0.038 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".