Machine-learning clustering analysis identifies novel phenogroups in patients with ST-elevation acute myocardial
Andrija Matetic1, Theocharis Kyriacou2, Mamas A Mamas3
1Department of Cardiology, University Hospital of Split, Split, Croatia; Keele Cardiovascular Research Group, Centre for Prognosis Research, Keele University, United Kingdom.
Insights
Machine learning identified four patient clusters for ST-elevation myocardial infarction (STEMI), revealing distinct risk profiles and outcomes. The "diabetes with end-organ damage" cluster showed the worst prognosis, while the "behavioral risk" cluster had the best outcomes.
Area of Science:
- Cardiovascular Medicine
- Artificial Intelligence in Healthcare
- Health Informatics
Background:
- Machine learning clustering can stratify patients with ST-elevation acute myocardial infarction (STEMI).
- Understanding patient subgroups is crucial for optimizing risk assessment, management strategies, and predicting prognosis in STEMI.
- Previous studies have not fully elucidated distinct phenogroups within the STEMI population using data-driven approaches.
Purpose of the Study:
- To apply machine learning clustering to identify distinct patient subgroups within STEMI.
- To compare the risk profiles, clinical outcomes, and management patterns across identified STEMI clusters.
- To investigate the prognostic implications of these data-driven phenogroups.
Main Methods:
- Utilized the National Inpatient Sample database (October 2015-December 2019) for adult STEMI discharges, excluding prior myocardial infarction cases.
- Employed machine-learning clustering based on 21 clinical attributes to define patient subgroups.
- Performed multivariable logistic regression to compare outcomes and management across clusters, calculating adjusted odds ratios (aOR) and 95% confidence intervals (95% CI).
Main Results:
- Identified four distinct STEMI clusters from 470,960 cases: 'behavioural risk' (n=205,640), 'least comorbidity' (n=146,400), 'diabetes with end-organ damage' (n=45,100), and 'cardiometabolic' (n=73,820).
- Hypertension and diabetes were the most significant attributes for clustering.
- The 'diabetes with end-organ damage' cluster had the highest mortality, MACCE, and ischemic stroke rates, with the lowest invasive management utilization (p < 0.001).
- The 'behavioural risk' cluster demonstrated the best in-hospital prognosis and highest invasive management rates (p < 0.001).
Conclusions:
- Machine learning effectively identifies distinct patient subgroups in STEMI, characterized by unique prevalence, risk profiles, prognoses, and management patterns.
- These data-driven phenogroups highlight significant variations in outcomes and treatment utilization among STEMI patients.
- Further research into these high-risk phenogroups is warranted to refine clinical decision-making and improve patient care.
Background:
Machine learning clustering of patients with ST-elevation acute myocardial infarction (STEMI) may provide important insights into their risk profile, management and prognosis.
Methods:
All adult discharges for STEMI in the National Inpatient Sample (October 2015 to December 2019) were included, excluding patients with prior myocardial infarction. Machine-learning clustering analysis was used to define clusters based on 21 clinical attributes of interest. Main outcomes of the study were cluster-based comparison of risk profile, in-hospital clinical outcomes and utilization of invasive management. Binomial hierarchical multivariable logistic regression with adjusted odds ratios (aOR) and 95% confidence intervals (95% CI) was used to detect the between-cluster differences.
Results:
Out of overall 470,960 STEMI cases, the machine-learning analysis revealed 4 different clusters with 205,640 (cluster 0: 'behavioural risk cluster'), 146,400 (cluster 1: 'least comorbidity cluster'), 45,100 (cluster 2: 'diabetes with end-organ damage cluster') and 73,820 (cluster 3: 'cardiometabolic cluster') cases. Attributes with the highest importance for clustering were hypertension and diabetes. After multivariable adjustment, patients from 'diabetes with end-organ damage cluster' exhibited the worst mortality, MACCE and ischemic stroke (p < 0.001 for all), as well as the lowest utilization of invasive management (p < 0.001 for all), in comparison to other clusters. Patients from 'behavioural risk cluster' exhibited the best in-hospital prognosis and the highest utilization of invasive management, compared to other clusters (p < 0.001 for all).
Conclusions:
Machine learning driven clustering of inpatients with STEMI reveals important population subgroups with distinct prevalence, risk profile, prognosis and management. Data driven approaches may identify high risk phenogroups and warrants further study.
More Related Videos
09:21Author Spotlight: Generating Neuronal Phenotypic Profiles - A Protocol to Culture and Image Human Midbrain Dopaminergic Neurons
Published on: July 7, 2023
08:51Author Spotlight: Integrated Multi-Omics Analysis for Unveiling Multicellular Immune Signatures in Clinical Heart Attack Cohorts
Published on: September 20, 2024
Related Concept Videos
01:27Myocarditis II: Clinical Features and Diagnostic Tests
01:30Acute Coronary Syndrome I: Introduction
01:19Acute Coronary Syndrome II: Pathophysiology and Clinical Manifestations
01:30Acute Coronary Syndrome III: Diagnostic Studies