Integration of polarization analytics and deep learning for a comprehensive experimental study of structural changes in biological tissues

Ryzhova, V.A., Khoteev A.A., Khlynov, R.D., Korotaev, V.V.

For Russian citation (Opticheskii Zhurnal):

Рыжова В.А., Хотеев А.А., Хлынов Р.Д., Коротаев В.В. Интеграция методов поляризационного анализа и глубокого обучения для комплексного экспериментального исследования структурных изменений в биологических тканях // Оптический журнал. 2025. Т. 92. № 12. С. 94–105. http://doi.org/10.17586/1023-5086-2025-92-12-94-105

Ryzhova V.A., Khoteev A.A., Khlynov R.D., Korotaev V.V. Integration of polarization analytics and deep learning for a comprehensive experimental study of structural changes in biological tissues [in Russian] // Opticheskii Zhurnal. 2025. V. 92. № 12. P. 94–105. http://doi.org/10.17586/1023-5086-2025-92-12-94-105

For citation (Journal of Optical Technology):

Abstract:

Subject of the study. Coordinate distributions of optical anisotropy parameters of biological tissues in polarization images. Aim of study. Development of a methodology that enables the integration of the results of statistical and correlation analysis of polarization images with deep learning methods for the classification of structural changes in optically heterogeneous tissues, including biological ones. Method. Physical modeling of laser radiation polarization transformation in biological tissues. Statistical analysis of coordinate distributions of the parameters of the anisotropic structure of the sample. Classification of images corresponding to changes in the structure of biological tissues using artificial neural networks. Main results. A model of a Mueller video polarimeter and a technique for forming and processing polarization images have been developed, allowing for the identification of informative parameters of the medium. It has been shown that the degree of polarization and dichroism are the most sensitive markers of structural changes in tissues. A technique for classifying biological tissues by changes in their structure has been developed using neural network models trained on datasets obtained from polarization images. Practical significance. The results of the work demonstrate the possibility of automated classification of biological tissues based on polarimetric data characterizing changes in the internal structure with an accuracy of up to 92%.

Keywords:

polarization image, Mueller matrix, structure of biological tissue, optical anisotropy, deep learning, neural network models

Acknowledgements:

the study was supported by the State funding allocated to the Pavlov Institute of Physiology Russian Academy of Sciences (№ 1021062411653-4-3.1.8). The work was supported by Project № 624083 Research work of master’s and postgraduate students of ITMO University

OCIS codes: 170.0170, 110.0110, 330.0330

References:

1. Guan C., Zeng N., He H. Review of polarization-based technology for biomedical applications // Journal of Innovative Optical Health Sciences. 2025. V. 18. № 2. P. 2430002. https://doi.org/10.1142/S1793545824300027
2. Pham T.T., Nguyen T.B., Sao Dam M. et al. A review of the application of the laser-light backscattering imaging technique to agricultural products // Agriculture. 2024. V. 14. № 10. P. 1782. https://doi.org/10.3390/agriculture14101782
3. Li Y., Zhuo Z., Liu C. et al. Deep learning enables accurate brain tissue microstructure analysis based on clinically feasible diffusion magnetic resonance imaging // NeuroImage. 2024. V. 300. P. 120858. https://doi.org/10.1016/j.neuroimage.2024.120858
4. Dong T., Lee H. H., Zang H. et al. In vivo cortical microstructure mapping using high-gradient diffusion mri accounting for intercompartmental water exchange effects // NeuroImage. 2025. P. 121258. https://doi.org/10.1016/j.neuroimage.2025.121258
5. Wu X., Pankow M., Onuma T. et al. Comparison of high-speed polarization imaging methods for biological tissues // Sensors. 2022. V. 22. № 20. P. 8000. https://doi.org/10.3390/s22208000
6. Li X., Yan L., Qi P. et al. Polarimetric imaging via deep learning: A review // Remote Sensing. 2023. V. 15. № 6. P. 1540. https://doi.org/10.3390/rs15061540
7. Liu Z., Song J., Fu Q. et al. Study on anisotropy orientation due to well-ordered fibrous biological microstructures // Journal of Biomedical Optics. 2024. V. 29. № 5. P. 052919–052919. https://doi.org/10.1117/1.JBO.29.5.052919
8. Lu S.Y., Chipman R.A. Interpretation of Mueller matrices based on polar decomposition // J. Opt. Soc. Am. A. 1996. Т. 13. № 5. С. 1106–1113. https://doi.org/10.1364/JOSAA.13.001106
9. Gros É., Rodríguez-Núñez O., Felger L. et al. Characterization of polarimetric properties in various brain tumor types using wide-field imaging Mueller polarimetry // IEEE transactions on medical imaging. 2024. V. 43. № 12. P. 4120–4132. https://doi.org/10.1109/TMI.2024.3413288
10. Ushenko A., Dubolazov A., Zheng J. et al. Mueller matrix polarization interferometry of optically anisotropic architectonics of biological tissue object fields: the fundamental and applied aspects // Frontiers in Physics. 2024. V. 11. P. 1302254. https://doi.org/10.3389/fphy.2023.1302254
11. Хлынов Р.Д., Рыжова В.А., Коротаев В.В. и др. Поляризационный неинвазивный метод мониторинга гематокрита крови // Оптический журнал. 2023. Т. 90. № 1. С. 60–75. https://doi.org/10.17586/1023-5086-2023-90-01-60-75
Khlynov R.D., Ryzhova V.A., Korotaev V.V. et al. Noninvasive polarization-based technique for hematocrit monitoring // Journal of Optical Technology. 2023. V. 90. № 1. P. 33–41. https://doi.org/10.1364/JOT.90.000033
12. Rodríguez C., Estévez I., González-Arnay E. et al. Optimizing the classification of biological tissues using machine learning models based on polarized data // Journal of Biophotonics. 2023. V. 16. № 4. P. e202200308. https://doi.org/10.1002/jbio.202200308
13. Le T.H., Huynh N.T., Phan Q.H. et al. Combination of Muller matrix imaging polarimetry and artificial intelligence for classification of mice skin cancer tissue in-vitro and in-vivo // Optik. 2024. V. 311. P. 171932. https://doi.org/10.1016/j.ijleo.2024.171932
14. Gil J.J., San José I. Mueller matrix associated with an arbitrary 4×4 real matrix. The effective component of a Mueller matrix // Photonics. 2025. V. 12. P. 230. https://doi.org/10.3390/photonics12030230
15. Zhang W.J. General correlation and partial correlation analysis in finding interactions: with Spearman rank correlation and proportion correlation as correlation measures // Network Biology. 2015. V. 5. № 4. P. 163. https://doi.org/10.0000/issn-2220-8879-networkbiology-2015-v5-0013
16. Sapkota R., Flores-Calero M., Qureshi R. et al. YOLO advances to its genesis: a decadal and comprehensive review of the You Only Look Once (YOLO) series // Artificial Intelligence Review. 2025. V. 58. № 9. P. 274. https://doi.org/10.1007/s10462-025-11253-3