Metabolic signatures and diagnostic models of ischemic stroke and its hypertensive subtype: a non-targeted metabolomics and machine learning.

Metabolic signatures of ischemic stroke and hypertensive subtype identified with diagnostic models achieving AUCs over 0.86 in validation sets.

• Why it matters: Reliable early diagnostic biomarkers for ischemic stroke are lacking, hindering timely intervention and treatment outcomes.
• What they did: Non-targeted metabolomics profiling of plasma samples from 100 patients and controls, combined with machine learning algorithms, identified differential metabolites and constructed diagnostic models.
• The result: Key metabolites L-Cystine, LPC 18:1, and LPC 14:0 demonstrated high diagnostic accuracy, validated through targeted metabolomics, supporting their potential as early biomarkers for IS and HIS.