Early Detection of Myocardial Infarction Based on Surface-Enhanced Raman Spectroscopy and Machine Learning

Abstract

This study addresses the problem of early diagnosis of myocardial infarction, one of the most common and life-threatening cardiovascular diseases. Early detection of the disease enables timely treatment, reduces the risk of mortality, and helps prevent severe complications. However, conventional clinical and laboratory diagnostic methods do not always provide reliable detection at the early stages of the disease. Therefore, the development of new minimally invasive and highly sensitive diagnostic approaches remains an important challenge.
In this work, a diagnostic approach based on surface-enhanced Raman spectroscopy (SERS), multivariate curve resolution–alternating least squares (MCR-ALS), and machine learning algorithms is proposed. The MCR-ALS method was used to extract informative spectral components, which were subsequently classified using logistic regression, support vector machine, random forest, and gradient boosting models.
The study included 85 blood serum samples, comprising 45 samples from patients with myocardial infarction and 40 samples from healthy controls. The diagnostic performance of the investigated models was evaluated and compared. The random forest model achieved the highest classification accuracy after MCR-ALS decomposition (Accuracy = 0.88 ± 0.08), while the logistic regression model demonstrated the highest area under the ROC curve (AUC = 0.96 ± 0.04).
The obtained results demonstrate that the combination of SERS, MCR-ALS, and machine learning algorithms enables effective discrimination between patients with myocardial infarction and healthy individuals. The proposed approach has the potential to serve as the basis for the development of rapid and minimally invasive systems for the early diagnosis of myocardial infarction.

Speaker

Pimenova Irina Aleksandrovna
Samara University
Russia

Discussion

Ask question