Key result
GBDT+SC improves prehospital ACS prediction sensitivity by ~8% alongside higher specificity.
Why the study?
Current machine learning approaches lack sufficient performance to safely rule-in or rule-out ACS in the prehospital setting.
Does an ensemble of gradient boosted decision trees with selective classification improve acute coronary syndrome prediction in patients transported by ambulance with chest pain?
Cohort (n=3,646)
No
Does an ensemble of gradient boosted decision trees with selective classification improve acute coronary syndrome prediction in patients transported by ambulance with chest pain?
Combining gradient boosted decision trees with selective classification improves the sensitivity and specificity of prehospital acute coronary syndrome prediction compared to current machine learning approaches.
May refine prehospital ACS triage; extends ML benchmarks but remains hypothesis-generating pending prospective validation.
Accurate identification of acute coronary syndrome (ACS) in the prehospital setting is important for timely treatments that reduce damage to the compromised myocardium. Current machine learning approaches lack sufficient performance to safely rule-in or rule-out ACS. Our goal is to identify a method that bridges this gap. To do so, we retrospectively evaluate two promising approaches, an ensemble of gradient boosted decision trees (GBDT) and selective classification (SC) on consecutive patients transported by ambulance to the ED with chest pain and/or anginal equivalents. On the task of ACS classification with 23 prehospital covariates, we found the fusion of the two (GBDT+SC) improves the best reported sensitivity and specificity by 8% and 23% respectively. Accordingly, GBDT+SC is safer than current machine learning approaches to rule-in and rule-out of ACS in the prehospital setting.
No takes yet. Share an insight, caveat, or question.
Garcia et al. (2026) conducted a cohort in Acute Coronary Syndrome (ACS) (n=3,646). Gradient Boosted Decision Trees with Selective Classification (GBDT+SC) vs. Current machine learning approaches (e.g., GBDT alone, Takeda, Al-Zaiti) and traditional HEART score was evaluated on Sensitivity for ACS classification on the external cohort. The fusion of gradient boosted decision trees and selective classification (GBDT+SC) improved the best reported sensitivity and specificity for prehospital ACS prediction by 8% and 23%, respectively.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: