CC BY
1*
ALT'23
The 30th International Conference on Advanced Laser Technologies
B-I-18
Laser spectroscopy biomedical data analysis and interpretation using
machine learning
Yury V. Kistenev, Alexey V. Borisov, Viktor E.Skiba, Vladimir V.Prishepa, Denis A. Vrazhnov
Laboratory of laser molecular imaging and machine learning, Tomsk State University, Tomsk 634050,
Russian Federation Corresponding author email: yuk@iao.ru
An idea of using exhaled air and a carrier of important information for medical diagnostics is very attractive because such diagnostic method is fast, noninvasive, comfortable for patients. But implementation of this idea requires applications of effective methods of spectral data analysis, including a preprocessing, informative features extraction, and predictive data model creation and validation. We plan to present results of application of machine learning methods [1], including deep learning [2] in every step of spectral data analysis pipeline presented above.
The research was carried out with the support of a grant under the Decree of the Government of the Russian Federation No. 220 of 09 April 2010 (Agreement No. 075-15-2021-615 of 04 June 2021).
References
[1] A. V. Borisov, A. G. Syrkina, D. A. Kuz'min, V. V. Ryabov, A. A. Boyko, O. Zaharova, V. S. Zasedatel', and Y. V. Kistenev, Application of machine learning and laser optical-acousticspectroscopy to study the profile of exhaled air volatile markers of acute myocardial infarction, J. Breath Res. 15 (2), 027104 (2021).
[2] Yu. V. Kistenev, V. E. Skiba, V. V. Prischepa, D. A. Vrazhnov, and A. V. Borisov, Super-resolution reconstruction of noisy gas-mixture absorption spectra using deep learning, J. Quant. Spectrosc. Radiat. Transfer (2022). https://doi.org/10.1016/jjqsrt.2022.108278
