Non‐Invasive Diagnosis of Hypertrophic Cardiomyopathy by Breath

• Published in: Advanced Sensor Research Vol. 4; no. 7
• Main Authors: Hershkovitz‐Pollak, Yael, Habib, Manhal, Broza, Yoav Y., Katz, Olga, Rakowski, Harry, Haick, Hossam
• Format: Journal Article
• Language: English
• Published: Wiley-VCH 01.07.2025
• Subjects: chromatography; hypertrophic cardiomyopathy (HCM); machine learning (ML); noninvasive; sensor; volatile organic compound (VOC)
• ISSN: 2751-1219; 2751-1219
• DOI: 10.1002/adsr.70012

Undetected in many patients, hypertrophic cardiomyopathy (HCM) is a widespread genetic heart disorder. Conventional diagnosis is based on physiological metrics such as blood pressure, imaging techniques, and genetic testing. Detection of HCM is crucial for proper follow‐up, family screening, early treatment, and risk stratification to prevent sudden cardiac death. Therefore, there is an unmet need for fast and reliable diagnostic methods. This study introduces an innovative approach for the noninvasive, rapid, and accurate diagnosis of HCM by detecting patterns of volatile organic compounds (VOCs) in the patient's breath. Breath from 157 volunteers is collected on sorbent tubes and analyzed using a two‐step approach, gas chromatography‐mass spectrometry (GC‐MS), and a developed nano‐based sensor array. Initially, statistically significant differences in VOC patterns among sampled groups are identified using GC‐MS. Then, the sensor array is used to differentiate between HCM patients and controls, resulting in the testing set, with 92.9% accuracy, 75% specificity, and 94.7% sensitivity. The sensors can further classify subcategories of HCM with >70.3% accuracy for all cases in the testing set. These findings show the applicability of the sensors setup and suggest that VOCs may be a promising noninvasive and real‐time HCM diagnosis option. The use of sensor technologies for the diagnosis of hypertrophic cardiomyopathy (HCM) by identification of volatile organic compounds (VOCs) in breath samples. By combining gas chromatography‐mass spectrometry (GC‐MS) and electronic nose (eNose) platforms, specific VOC patterns associated with HCM are detected and analyzed. Results reveal distinct metabolic signatures, supporting sensor‐based, non‐invasive diagnostic approaches for early and accurate detection of HCM.