Machine-learning facilitates selection of a novel diagnostic panel of metabolites for the detection of heart failure

Abstract

the metabolic derangement is common in heart failure with reduced ejection fraction (Hfref). the aim of the study was to check feasibility of the combined approach of untargeted metabolomics and machine learning to create a simple and potentially clinically useful diagnostic panel for Hfref. the study included 67 chronic HFrEF patients (left ventricular ejection fraction-LVEF 24.3 ± 5.9%) and 39 controls without the disease. Fasting serum samples were fingerprinted by liquid chromatographymass spectrometry. Feature selection based on random-forest models fitted to resampled data and followed by linear modelling, resulted in selection of eight metabolites (uric acid, two isomers of Lpc 18:2, LPC 20:1, deoxycholic acid, docosahexaenoic acid and one unknown metabolite), demonstrating their predictive value in Hfref. the accuracy of a model based on metabolites panel was comparable to BNP (0.85 vs 0.82), as verified on the test set. Selected metabolites correlated with clinical, echocardiographic and functional parameters. the combination of two innovative tools (metabolomics and machine-learning methods), both unrestrained by the gaps in the current knowledge, enables identification of a novel diagnostic panel. Its diagnostic value seems to be comparable to BNP. Large scale, multi-center studies using validated targeted methods are crucial to confirm clinical utility of proposed markers.

Metabolomic Profiling of Adults with Congenital Heart Disease

“…In this scenario, the integrative capacity of metabolomic analysis has the potential to uncover unifying and potentially targetable metabolic phenotypes generated by a broad diversity of upstream systemic inputs. In support of this hypothesis, the pairing of metabolomic analysis with computer modeling has recently yielded novel insight into the etiologically heterogeneous pathophysiology of acquired systolic heart failure [ 14 ].…”

Artificial Intelligence and Cardiovascular Risk Prediction: All That Glitters is not Gold

“…The accuracy of the model based on the eight selected metabolites did not outperform BNP alone in the prediction of HFrEF (AUC 0.82 versus 0.85), but the study demonstrated that the combination of metabolomics and machinelearning methods enables the identification of novel diagnostic panels. 29 Machine learning models and multiple risk calculation scoring systems Using characteristics from electronic health records of 1,560 patients with HFrEF, the authors trained their model using different algorithms (random forest, logistic regression, support vector regression, decision tree and AdaBoost). Compared with the SHFM, their model resulted in improved performance for predicting 1-, 2-, and 5-year survival, with an improvement in the AUC of 11%; of note, logistic regression and random forest were the most accurate approaches.…”

The Evolution and Future Direction of The Cardiac Biomarker

“…Increases or decreases of >30 miRNA may orchestrate changes to the transcriptome, and ultimately the proteome, during HF. This may, for example, lead to differentially expressed proteins involved in glycolysis, β-oxidation, and ketone metabolism in the failing heart, 72 as well as promote the development of cardiac fibrosis, e.g., as caused by miRNA-21.…”