Abstract 17121: Predictors of Death and Major Adverse Cardiovascular Events in the ACCORD Trial Identified by Random Survival Forest Based Machine-Learning
Bibliographic record
Abstract
Background: Patients with type 2 diabetes (T2D) are at high risk of cardiovascular (CV) morbidity/mortality. The ACCORD trial (NCT00000620) tested intensive glycemic, lipid and blood pressure interventions on major CV events in 10,251 T2D patients with baseline HbA1c concentration >7.5%. Despite its landmark findings, a data-driven systematic evaluation of predictors for major cardiovascular events among hundreds of ACCORD variables has not been conducted. Methods: Random Survival Forest (RSF), a machine-learning method for survival analysis, identified important predictors for total mortality (TM), CV death (CVd), hospitalization/death due to heart failure (hdHF), fatal/non-fatal stroke (CVA), non-fatal myocardial infarction (MI) and MACE (composite of CVd, MI and CVA). Among 378 risk factors, including some highly correlated features, the top-ranked predictors (collected at baseline or derived from repeated measures prior to events) were selected, resulting in a hierarchy of predictive variables. Effects of RSF-selected predictors were then evaluated by multivariate Cox Proportional Hazards Models. Results: Table 1 presented the top ten predictors for six major events. Variables associated with changes in renal function predicted TM, CVd, and hdHF with ~90% accuracy. Insulin use was an important predictor along with predefined composite renal microvascular events for MI, CVA and MACE (74-79% accuracy). The Cox regression models based on RSF variable selection yielded similar findings for these important predictors of events. Conclusions: RSF approach revealed that insulin use and overt renal microvascular events were predictors for the occurrence of MI, stroke, and MACE in T2D patients. Moreover, dynamic changes in urinary renal function biomarkers had additional predictive values for fatal events. These results provide important clinical insights for reducing CV events in Type 2 diabetes patients.
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 | 0.007 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".