Decoding Diabetes: Harnessing AI to Accurately Predict Real-Time and Future Blood Glucose Levels for Diabetes Management Using Diet, Exercise, Insulin Intake, and Heart Rate Variability
Bibliographic record
Abstract
Abstract Continuous glucose monitoring (CGM) systems play a crucial role in diabetes care. Yet, they focus solely on blood glucose levels (BGL), neglect diet, exercise, and medication, and lack predictive capabilities, leaving patients and clinicians with reactive rather than proactive solutions. This study introduces Dual Temporal Recurrent Ensemble (DTRE), a novel AI model that bridges these gaps by enabling real-time BGL monitoring without a traditional CGM and forecasting BGL for 30 to 120 minutes into the future when integrated with CGM data. The model’s performance is based on two parallel branches of hybrid models that combine advanced machine learning architectures for accurate predictions and integrate key biomarkers like diet, exercise, insulin intake, heart rate (HR), and heart rate variability (HRV). This pioneering study developed and validated DTRE on two diverse datasets, OHIOT1DM and D1NAMO, achieving exceptional forecasting accuracy: a Mean Absolute Relative Difference (MARD) of 7.6% at 30 minutes and 19.2% at 120 minutes, outperforming existing models by 13 to 41%. DTRE also surpassed commercial CGM systems for real-time BGL predictions by achieving a MARD of 7.17% on the high-frequency D1NAMO dataset compared to FreeStyle Libre 3 (7.9%) and Dexcom G7 (8.2%). DTRE is the first AI-driven virtual BGM (vBGM) to integrate all four pillars of diabetes care and validate them on two diverse datasets. It offers a non-invasive, low-cost, and proactive solution. Its real-time insights can provide patients and clinicians with actionable data, transforming diabetes management for individuals who depend on insulin. Validation across diverse datasets underscores its potential for global application. Human Subject Research Statement This research did not utilize any human subjects. Pre-existing and published databases were used for this research.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".