Integrating untargeted metabolomics and machine learning to reveal an aberration of sphingolipid metabolism in cardiometabolic HFpEF.
Metabolomics identifies a specific fat molecule that distinguishes a type of heart failure from similar metabolic disease, potentially aiding diagnosis.
This study integrates untargeted metabolomics with machine learning to characterize the metabolic fingerprint distinguishing cardiometabolic HFpEF from metabolic syndrome without HF, identifying sphingolipid dysregulation—particularly C24:1 sphingomyelin—as a potential diagnostic biomarker associated with NT-proBNP. While preliminary, the independent cohort validation of total SM levels provides initial external support for sphingomyelin's clinical applicability in this challenging-to-diagnose population.
What the study was
- Study design
- metabolomics_ml_pilot
- Population
- Cardiometabolic HFpEF vs metabolic syndrome patients
- Category
- Diagnostics
- Maturity
- Exploratory
- Journal
- Diabetology & Metabolic Syndrome
Why it surfaced
Cardiometabolic HFpEF is a high-prevalence, difficult-to-diagnose phenotype; sphingomyelin as a diagnostic biomarker is a novel lead; SHAP-based feature attribution provides methodological transparency; preliminary but externally confirmed signal warrants follow-up.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.