Volatilome and machine learning in ischemic heart disease: Current challenges and future perspectives

Abstract

Integrating exhaled breath analysis into the diagnosis of cardiovascular diseases holds significant promise as a valuable tool for future clinical use, particularly for ischemic heart disease (IHD). However, current research on the volatilome (exhaled breath composition) in heart disease remains underexplored and lacks sufficient evidence to confirm its clinical validity. Key challenges hindering the application of breath analysis in diagnosing IHD include the scarcity of studies (only three published papers to date), substantial methodological bias in two of these studies, and the absence of standardized protocols for clinical implementation. Additionally, inconsistencies in methodologies—such as sample collection, analytical techniques, machine learning (ML) approaches, and result interpretation—vary widely across studies, further complicating their reproducibility and comparability. To address these gaps, there is an urgent need to establish unified guidelines that define best practices for breath sample collection, data analysis, ML integration, and biomarker annotation. Until these challenges are systematically resolved, the widespread adoption of exhaled breath analysis as a reliable diagnostic tool for IHD remains a distant goal rather than an imminent reality.

Keywords: Volatilome; Breathome; Ischemic heart disease; Mass spectrometer; Machine learning

Core Tip: Exhaled breath analysis offers a non-invasive, cost-effective alternative to traditional ischemic heart disease diagnostics, with superior accuracy (84% vs 60%–70% for stress electrocardiography). To enhance reliability, standardized protocols for breath collection and the integration of machine learning are essential. Collaborative efforts among clinicians, chemists, and data scientists are key to unlocking its full clinical potential.