Two schemes for interpolation of optimal values of ventilation flow parameters depending on the values of patient indicators during artificial ventilation of the lungs

This article, based on a database of initial data on successful patient treatment, proposes two schemes for interpolating optimal ventilation flow parameter values during artificial lung ventilation (ALV) for a given patient. At the mathematical level, selecting optimal ventilation flow parameter values based on the patient's current condition is a task of multivariate nonlinear regression analysis. The first scheme is based on the mathematical apparatus of artificial neural networks. The second scheme is based on the mathematical apparatus of metric analysis, developed at the Department of Applied Mathematics at MEPhI and currently used in mathematical data processing and optimization problems in various applied fields. The implementation of both schemes allows for the use of accumulated data on the successful treatment of patients with similar lung diseases on ventilators for the specific patient in question. Both schemes allow for the adaptation of optimal ventilation flow parameter values to the patient's current condition during treatment. In the future, it is planned to jointly use these two interpolation schemes to obtain a more accurate and reliable final result for solving the above-mentioned optimal interpolation problem.

1. Chursin V.V. Iskusstvennaya ventilyaciya lyogkih: Uchebno-metodicheskoe posobie. [Artificial ventilation of the lungs: Textbook and methodological manual]. Almaty, 2008. 55 p. ISBN 9965-874-64-6.

2. Kuzkov V.V., Suborov E.V., Fot E.V., Rodionova L.N., Sokolova M.M., Lebedinskij K.M., Kirov M.Yu. Posle-operacionnye dyhatel'nye oslozhneniya i ORDS legche predupredit', chem lechit' [Post-operative respiratory complications and ARDS are easier to prevent than to treat]. Anesteziologiya i reanimatologiya, 2016. No.61(6). Pp.461-468. (in Russian)

3. Patent RU2003121722A. Sposob provedeniya dlitel'noj iskusstvennoj ventilyacii legkih [Method for Performing Long-Term Artificial Lung Ventilation]. Google Patents. 2019. https://patents.google.com/patent/RU2003121722A/ru

4. Li M., Jiang Y., Zhang Y., Zhu H. Medical image analysis using deep learning algorithms. Frontiers in Public Health,2023. Vol.11. 1273253. DOI: 10.3389/fpubh.2023.1273253.

5. Zhang L., Zhu E., Shi J., Wu X., Cao S., Huang S., Ai Z., Su J. Individualized treatment recommendations for patients with locally advanced head and neck squamous cell carcinoma utilizing deep learning. Front. Med., 2025. Vol.11. 1478842. DOI: 10.3389/fmed.2024.1478842.

6. Perkins S. W., Muste J. C., Alam T., Singh R. P. Improving Clinical Documentation with Artificial Intelligence: A Systematic Review. Perspectives in health information management, 2024. Vol.21(2), 1d. PMCID: PMC11605373

7. Kryanev A.V., Lukin G.V., Udumyan D.K. Metricheskij analiz i obrabotka dannyh [Metric analysis and data processing] Moscow, Fizmatlit Publ., 2012. 308 p.

8. Kryanev A.V., Udumyan D.K. Metric analysis, properties and applications as a tool for interpolation. International Journal of Mathematical Analysis, 2014. Vol. 8 (45). Pp. 2221-2228. DOI:10.12988/ijma.2014.48252

9. Kryanev A.V., Udumyan D. K. Metric Analysis, Properties and Applications as a Tool for Forecasting. International Journal of Mathematical Analysis, 2014. Vol. 8. No. 60. Pp. 2971 – 2978. DOI: 10.12988/ijma.2014.411341

10. Ivanov V.V. Kryanev A.V., Udumyan D.K., Lukin G.V. Metric Analysis Approach for Interpolation and Forecasting of Time Processes. Applied Mathematical Sciences, 2014. Vol. 8. No. 22. Pp. 1053 – 1060. DOI: 10.12988/ams.2014.312727

11. Klimanov S.G., Kotlyarov A.A., Kryanev A.V., et al. Sravnenie metodov vyyavleniya anomal'nyh vybrosov v iskhodnyh dannyh i ih primenenie pri obrabotke dannyh iskusstvennoj ventilyacii legkih [Comparison of methods for identifying abnormal outliers in source data and their application in processing artificial lung ventilation data]. Vestnik NIYaU MIFI, 2025. Vol.14(1). Pp.37-49. (in Russian) DOI: 10.26583/vestnik.2025.1.4.

12. Tavoosi J., Zhang Ch., Mohammadzadeh A., Mobayen S., Mosavi A. H. Medical Image Interpolation Using Recurrent Type-2 Fuzzy Neural Networks. Frontiers in Neuroinformatics, 2021. Vol.15. 667375. DOI: 10.3389/fninf.2021.667375

13. Vaganov S. E. Adaptivnyj nejrosetevoj metod postroeniya interpolyacionnoj formuly dlya udvoeniya razmera izobrazheniya [An adaptive neural network method for constructing an interpolation formula for doubling the image size]. Komp'yuternaya optika, 2019. Vol. 43, no. 4. pp. 627-631. DOI: 10.18287/2412-6179-2019-43-4-627-631.

14. Gambini L., Gabbett C., Doolan L., Jones L., et al. Video frame interpolation neural network for 3D tomography across different length scales. Nature Communications, 2024. Vol.15 (1). 7962. DOI: 10.1038/s41467-024-52260-2.

15. Hariharan S., Karnan H., Maheswari D.U. Automated mechanical ventilator design and analysis using neural network. Scientific Reports, 2025. Vol.15. 3212. DOI: 10.1038/s41598-025-87946-0.

16. Peine A., Hallawa A., Bickenbach J., et al. Development and validation of a reinforcement learning algorithm to dynamically optimize mechanical ventilation in critical care. NPJ Digital Medicine, 2021. Vol.4 (1). DOI: 10.1038/s41746-021-00388-6.

17. Diao S. Changsong W., Junyu W., Yizhou Li. Ventilator pressure prediction using recurrent neural network. DOI: 10.48550/arXiv.2410.06552.

18. Bakkes T., Diepen A. van, Bie A.De., Montenij L. Automated detection and classification of patient–ventilator asynchrony by means of machine learning and simulated data. Computer Methods and Programs in Biomedicine, 2023. Vol.230 (6). 107333 . DOI: 10.1016/j.cmpb.2022.107333.