Application of machine learning and laser optical-acoustic spectroscopy to study the profile of exhaled air volatile markers of acute myocardial infarction.

Borisov, Alexey; Syrkina, Anna; Kuzmin, Dmitry; Ryabov, Vyacheslav; Boyko, Andrey Aleksandrovich; Zaharova, Olga; Zasedatel, Vyacheslav; Kistenev, Yury

Conventional acute myocardial infarction (AMI) diagnosis is quite accurate and has proved its effectiveness. However, despite this, discovering more operative methods of this disease detection is underway. From this point of view, the application of exhaled air analysis for similar diagnosis is valuable. This paper aims to find effective machine learning algorithms to construct the predictive model for AMI diagnosis using experimental data of a patient's exhaled air absorption spectra. The target group included 30 patients with primary myocardial infarction. The control group included 42 healthy volunteers. The "LaserBreeze" laser gas analyzer (Special Technologies Ltd, Russia), based on the dual-channel resonant photoacoustic detector cell and optical parametric oscillator as the laser source, had been used. The informative features were extracted with the Principal Component Analysis. The predictive model was created using the Support Vector Machine, which provides 0.86 of the mean values of both sensitivity and specificity. The spectral subrange from 2.775 µm to 3.025 µm was shown to be preferable for acute myocardial infarction diagnosis using exhaled air spectral analysis.