Development of a combined neural network model for effective spectroscopic analysis

Authors

DOI:

https://doi.org/10.15587/1729-4061.2025.322627

Keywords:

deep learning models, numerical modeling, optimization method, spectral analysis, signal processing

Abstract

The object of this study is spectroscopic data from chemical, organic compounds, and physical experiments, characterized by signal complexity, low signal-to-noise ratio, and significant variability of acquisition conditions. The problem addressed is to improve the accuracy and stability of spectral data analysis in the tasks of identification and quantification of components, in particular under conditions of noise, variable baselines, and experimental parameters. The essence of the results is the designed optimized complex neural network model (CNN+LSTM), which provides high resistance to noise and variability of experimental parameters. The constructed neural network model of spectral analysis achieved concentration prediction accuracy at the level of R2=0.98 with RMSE less than 5 %, which significantly exceeds conventional methods. The implementation includes the use of modern optimizers for stable learning and software implementation in Python using the TensorFlow/Keras libraries. The features and differences that made it possible to solve the problem under consideration include development of the algorithm of automatic normalization of spectra, construction of synthetic training data set, adaptation of the model to low signal-to-noise ratio and resistance to changes under experimental conditions. The results are explained by the ability of the proposed neural network architecture to model nonlinear dependences, automatically allocate relevant features, and compensate for noise effects, which is critical for working with spectral data. The conditions of use in practice include pharmaceutical analysis tasks, environmental monitoring, physical and chemical analysis of complex multicomponent systems, especially with limited experimental resources and variable external factors

Author Biographies

Yurii Bilak, Uzhhorod National University

PhD, Associate Professor

Department of Software Systems

Antonina Reblian, Uzhhorod National University

Lecturer

Department of Software Systems

Roman Buchuk, Uzhhorod National University

PhD

Department of Software Systems

Pavlo Fedorka, Uzhhorod National University

PhD (in Philosophy), PhD

Department of Software Systems

References

  1. Khayatzadeh Mahani, M., Chaloosi, M., Ghanadi Maragheh, M., Khanchi, A. R., Afzali, D. (2007). Comparison of Artificial Neural Networks with Partial Least Squares Regression for Simultaneous Determinations by ICP‐AES. Chinese Journal of Chemistry, 25 (11), 1658–1662. https://doi.org/10.1002/cjoc.200790306
  2. Liu, X., An, H., Cai, W., Shao, X. (2024). Deep learning in spectral analysis: Modeling and imaging. TrAC Trends in Analytical Chemistry, 172, 117612. https://doi.org/10.1016/j.trac.2024.117612
  3. Primrose, M., Giblin, J., Smith, C., Anguita, M., Weedon, G. (2022). One dimensional convolutional neural networks for spectral analysis. Algorithms, Technologies, and Applications for Multispectral and Hyperspectral Imaging XXVIII, 12. https://doi.org/10.1117/12.2618487
  4. Schuetzke, J., Szymanski, N. J., Reischl, M. (2023). Validating neural networks for spectroscopic classification on a universal synthetic dataset. Npj Computational Materials, 9 (1). https://doi.org/10.1038/s41524-023-01055-y
  5. Liu, J., Osadchy, M., Ashton, L., Foster, M., Solomon, C. J., Gibson, S. J. (2017). Deep convolutional neural networks for Raman spectrum recognition: a unified solution. The Analyst, 142 (21), 4067–4074. https://doi.org/10.1039/c7an01371j
  6. Marini, F., Bucci, R., Magrì, A. L., Magrì, A. D. (2008). Artificial neural networks in chemometrics: History, examples and perspectives. Microchemical Journal, 88 (2), 178–185. https://doi.org/10.1016/j.microc.2007.11.008
  7. Butler, K. T., Davies, D. W., Cartwright, H., Isayev, O., Walsh, A. (2018). Machine learning for molecular and materials science. Nature, 559 (7715), 547–555. https://doi.org/10.1038/s41586-018-0337-2
  8. Mishra, P., Passos, D., Marini, F., Xu, J., Amigo, J. M., Gowen, A. A. et al. (2022). Deep learning for near-infrared spectral data modelling: Hypes and benefits. TrAC Trends in Analytical Chemistry, 157, 116804. https://doi.org/10.1016/j.trac.2022.116804
  9. Saravanan, N., Ganesan, M. (2024). Spectral Analysis of Cellular Neural Network: Unveiling Network Parameters and Graph Characteristics. https://doi.org/10.21203/rs.3.rs-4338706/v1
  10. Madden, M. G., Ryder, A. G. (2003). Machine learning methods for quantitative analysis of Raman spectroscopy data. Opto-Ireland 2002: Optics and Photonics Technologies and Applications, 4876, 1130. https://doi.org/10.1117/12.464039
  11. Randolph, T. W. (2006). Scale-based normalization of spectral data. Cancer Biomarkers, 2 (3-4), 135–144. https://doi.org/10.3233/cbm-2006-23-405
  12. Wei, J., Pan, S., Gao, W., Zhao, T. (2022). A dynamic object filtering approach based on object detection and geometric constraint between frames. IET Image Processing, 16 (6), 1636–1647. https://doi.org/10.1049/ipr2.12436
  13. Wiatowski, T., Bolcskei, H. (2018). A Mathematical Theory of Deep Convolutional Neural Networks for Feature Extraction. IEEE Transactions on Information Theory, 64 (3), 1845–1866. https://doi.org/10.1109/tit.2017.2776228
  14. Zhang, Z. (2018). Improved Adam Optimizer for Deep Neural Networks. 2018 IEEE/ACM 26th International Symposium on Quality of Service (IWQoS), 1–2. https://doi.org/10.1109/iwqos.2018.8624183
  15. Zaheer, R., Shaziya, H. (2019). A Study of the Optimization Algorithms in Deep Learning. 2019 Third International Conference on Inventive Systems and Control (ICISC). https://doi.org/10.1109/icisc44355.2019.9036442
  16. Shao, X., Bian, X., Liu, J., Zhang, M., Cai, W. (2010). Multivariate calibration methods in near infrared spectroscopic analysis. Analytical Methods, 2 (11), 1662. https://doi.org/10.1039/c0ay00421a
  17. Alsmeyer, F., Marquardt, W. (2004). Automatic Generation of Peak-Shaped Models. Applied Spectroscopy, 58 (8), 986–994. https://doi.org/10.1366/0003702041655421
  18. Hrytsak, R., Shuaibov, O., Minya, O., Malinina, A., Shevera, I., Bilak, Y., Homoki, Z. (2024). Conditions for pulsed gas-discharge synthesis of thin tungsten oxide films from a plasma mixture of air with tungsten vapors. Physics and Chemistry of Solid State, 25 (4), 684–688. https://doi.org/10.15330/pcss.25.4.684-688
  19. Shuaibov, O. K., Hrytsak, R. V., Minya, O. I., Malinina, A. A., Bilak, Yu. Yu., Gomoki, Z. T. (2022). Spectroscopic diagnostics of overstressed nanosecond discharge plasma between zinc electrodes in air and nitrogen. Journal of Physical Studies, 26 (2). https://doi.org/10.30970/jps.26.2501
  20. Bondar, I. I., Suran, V. V., Minya, O. Y., Shuaibov, O. K., Bilak, Yu. Yu., Shevera, I. V. et al. (2023). Synthesis of Surface Structures during Laser-Stimulated Evaporation of a Copper Sulfate Solution in Distilled Water. Ukrainian Journal of Physics, 68 (2), 138. https://doi.org/10.15407/ujpe68.2.138
  21. Kozubovsky, V. R., Bilak, Yu. Yu. (2022). Express Analysis of Gas Mixtures Using a Spectral Correlator Based on the Fabry–Perot Interferometer. Journal of Applied Spectroscopy, 89 (3), 495–499. https://doi.org/10.1007/s10812-022-01385-7
  22. Kozubovsky, V., Bilak, Y. (2021). Phase Methods in Absorption Spectroscopy. Ukrainian Journal of Physics, 66 (8), 664. https://doi.org/10.15407/ujpe66.8.664
Development of a combined neural network model for effective spectroscopic analysis

Downloads

Published

2025-02-24

How to Cite

Bilak, Y., Reblian, A., Buchuk, R., & Fedorka, P. (2025). Development of a combined neural network model for effective spectroscopic analysis. Eastern-European Journal of Enterprise Technologies, 1(4 (133), 41–51. https://doi.org/10.15587/1729-4061.2025.322627

Issue

Vol. 1 No. 4 (133) (2025): Mathematics and Cybernetics - applied aspects

Section

Mathematics and Cybernetics - applied aspects

License

Copyright (c) 2025 Yurii Bilak, Antonina Reblian, Roman Buchuk, Pavlo Fedorka

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

The consolidation and conditions for the transfer of copyright (identification of authorship) is carried out in the License Agreement. In particular, the authors reserve the right to the authorship of their manuscript and transfer the first publication of this work to the journal under the terms of the Creative Commons CC BY license. At the same time, they have the right to conclude on their own additional agreements concerning the non-exclusive distribution of the work in the form in which it was published by this journal, but provided that the link to the first publication of the article in this journal is preserved.

A license agreement is a document in which the author warrants that he/she owns all copyright for the work (manuscript, article, etc.).
The authors, signing the License Agreement with TECHNOLOGY CENTER PC, have all rights to the further use of their work, provided that they link to our edition in which the work was published.
According to the terms of the License Agreement, the Publisher TECHNOLOGY CENTER PC does not take away your copyrights and receives permission from the authors to use and dissemination of the publication through the world's scientific resources (own electronic resources, scientometric databases, repositories, libraries, etc.).
In the absence of a signed License Agreement or in the absence of this agreement of identifiers allowing to identify the identity of the author, the editors have no right to work with the manuscript.
It is important to remember that there is another type of agreement between authors and publishers – when copyright is transferred from the authors to the publisher. In this case, the authors lose ownership of their work and may not use it in any way.