Plasma metabolite model using 5 metabolites achieved 97.1% accuracy and AUROC 0.985 (SVM) and 100% accuracy and AUROC 1.000 (RF) in diagnosing HCM versus controls.
Case-Control (n=111)
No
Does a plasma metabolite diagnostic model using machine learning accurately identify patients with hypertrophic cardiomyopathy compared to healthy controls?
A machine learning diagnostic model utilizing five specific plasma metabolites demonstrated high accuracy in distinguishing patients with hypertrophic cardiomyopathy from healthy controls.
Effect estimate: SVM model AUROC 0.985 in validation set; accuracy 97.1%; RF model AUROC 1.000 in validation set; accuracy 100.0%
p-value: p=<0.05
Plasma metabolite diagnostic model including KAPA, γ-linolenoyl ethanolamid, nitrilotriacetic acid, D-quinovose and NAA can effectively and accurately screen HCM patients. Metabolomics combined with ML algorithm showed that alanine, aspartate and glutamate metabolism may be the pathogenic pathway leading to the occurrence of HCM with NAA as the central target.
Li et al. (Sat,) conducted a case-control in Adult patients (>18 years) with definitively diagnosed hypertrophic cardiomyopathy (HCM) per guidelines compared to healthy controls without cardiovascular or metabolic disease (n=111). Plasma metabolite diagnostic model including 7-keto-8-aminopelargonic acid (KAPA), γ-linolenoyl ethanolamid, nitrilotriacetic acid, D-quinovose and N-acetyl-l-aspartic acid (NAA) vs. Normal participants without cardiovascular or metabolic disease was evaluated on Accuracy and AUROC of diagnostic model differentiating HCM patients from controls (SVM model AUROC 0.985 in validation set; accuracy 97.1%; RF model AUROC 1.000 in validation set; accuracy 100.0%, p=<0.05). Plasma metabolite model using 5 metabolites achieved 97.1% accuracy and AUROC 0.985 (SVM) and 100% accuracy and AUROC 1.000 (RF) in diagnosing HCM versus controls.