Key result
Eight-variable model detects carotid plaque in T2DM with ~0.83 AUC but lacks external transportability.
Why the study?
The study aimed to develop and evaluate a parsimonious clinical model for detecting prevalent carotid plaque in patients with T2DM, and to assess its interpretability, robustness, and transportability.
Does an eight-variable logistic regression clinical model accurately detect prevalent carotid plaque in patients with type 2 diabetes mellitus compared to machine learning models?
Cross-Sectional (n=5,226)
Does an eight-variable logistic regression clinical model accurately detect prevalent carotid plaque in patients with type 2 diabetes mellitus compared to machine learning models?
An eight-variable logistic regression model performed comparably to machine learning models for detecting carotid plaque in hospitalized T2DM patients, but demonstrated unstable variable selection, confounding by indication, and poor external transportability.
Answers are grounded in this paper and cite their sources.
Clinicians should not adopt carotid plaque models without external validation; leaves open whether routine variables can achieve transportability in T2DM.
To develop and evaluate a parsimonious model for detecting prevalent carotid plaque in type 2 diabetes mellitus (T2DM), and to assess its interpretability, robustness, and transportability. This retrospective cross-sectional study enrolled 2,035 hospitalized patients with T2DM, randomly assigned 70:30 to training and test sets. Bidirectional stepwise regression screened 39 candidate variables; stability was assessed by bootstrap resampling (1,000 replicates) and penalized regression. Performance was compared with random forest and XGBoost, and TreeSHAP was used for interpretability. Interaction terms, a non-medication model, and high-sensitivity thresholds were evaluated. The model was applied to an independent cohort of 3,191 health-screening individuals, with updating assessed by split-half validation. Eight predictors were retained. The test-set AUC was 0.827 (95% CI 0.791–0.862), versus 0.804 for random forest and 0.816 for XGBoost ( P = 0.023 versus random forest). Only age and statin use were selected in 100% of bootstrap resamples. Removing statin use and GLP-1RA lowered the AUC to 0.787 (ΔAUC − 0.040, P < 0.001), retaining a history of coronary heart disease. No interaction term was significant (joint P = 0.511). In the external cohort, the AUC was 0.666 with marked miscalibration; updating left discrimination unchanged, and at 85% sensitivity, specificity was 33.4%. An eight-variable logistic model using routinely available variables performed comparably to machine learning models for detecting prevalent carotid plaque in the development cohort. However, variable selection was unstable, and statin use was the strongest but indication-confounded correlate. External discrimination was moderate and did not improve after recalibration, indicating limited transportability. The model is best regarded as an exploratory candidate requiring further setting-specific validation, rather than one that may already aid in the detection of clinical cases.
No takes yet. Share an insight, caveat, or question.
Liu et al. (2026) conducted a cross-sectional in Type 2 diabetes mellitus and carotid plaque (n=5,226). Eight-variable clinical logistic regression model vs. Random forest and XGBoost machine learning models was evaluated on Detection of prevalent carotid plaque (AUC) (AUC 0.827, 95% CI 0.791-0.862, p=0.023 versus random forest). An eight-variable logistic model detected prevalent carotid plaque in T2DM with a test-set AUC of 0.827, but showed limited external transportability.
Synapse has enriched 10 related papers. Consider them for comparative context: