Cardiovascular disease (CVD) remains a major global health concern and a leading cause of morbidity and mortality worldwide. Early-diagnosis and prompt medical attention are crucial in managing and reducing overall impact on health-and-wellbeing, necessitating the development of innovative diagnostics, which transcend traditional methodologies. Raman spectroscopy uniquely provides molecular fingerprinting and structural information, offering insights into biochemical composition. Integration of Raman spectroscopy with advanced machine learning is established as a powerful clinical adjunct for point-of-care detection of CVDs. A non-invasive, label-free spectroscopic platform coupled with neural network algorithm, 'SKiNET' has been developed to accurately detect the biomolecular changes within plasma of CVD versus healthy cohorts, enabling rapid diagnosis and longer-term monitoring, where the real-time capabilities provide dynamic assessment of progression, aligning treatment strategies with evolving states. CVD has been detected and classified via SKiNET with 88.6 %-accuracy, 92.9 %-specificity and 85.1 %-sensitivity and with 83.8 %-accuracy. The hybrid RS-SKiNET bio-molecularly specific detection signposted a comprehensive panel of CVD-indicative biomarkers, including SIL-6, IL-9, LpA, ApoB, PCSK9 and NT-ProBNP, offering important insights into disease mechanisms and risk-stratification. This multidimensional technique holds potential for improved patient-and-healthcare management for CVDs, laying the platform toward high-throughput biomolecular profiling of CVD-indicative macromolecular biomarkers, particularly vital for widespread point-of-care diagnostics and monitoring.
Keywords: Biomolecular spectroscopic profiling; Cardiovascular disease macromolecular markers; Label-free spectroscopy-AI technique; Non-invasive.
Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.