Збірник праць конференції «International Conference on Advanced Laser Technologies (ALT)»
Наукова стаття на тему 'Lymphedema tissue analysis using optical imaging and machine learning'

Текст наукової роботи на тему «Lymphedema tissue analysis using optical imaging and machine learning»


Lymphedema tissue analysis using optical imaging and machine learning

Y. Kistenev1, A. Borisov1, V. Nikolaev1, D. Vrazhnov1,2, A. Knyazkova1, N. Kryvova1, E. Sandykova3

1Tomsk State Unuversity, Physics, Tomsk, Russian Federation

2Institute of Strength Physics and Materials Science of Siberian Branch of the RAS, Tomsk,

Russia, Lab. of Phoacoustics, Tomsk, Russian Federation

3Siberian State Medical University, Phesics, Tomsk, Russian Federation

The term "lymphedema" is a kind of fat disorder caused by weak lymph infiltration in a tissue, leading to inflammation, hypertrophy of adipose tissue and fibrosis. The pathophysiology of lipedema is not clearly understood Lymphedema is divided into primary and secondary. The origin of primary lymphedema mostly connected with genomic abnormality. One of the most important reasons of the secondary lymphedema is complication after cancer surgery or radiotherapy treatment.

There is no cure for lipedema, but there are treatment approaches that allow to slow down the progression. Proper diagnosis and treatment will help prevent complications [1]. Various methods for lymphedema diagnosis are used at the stage of clinical manifestations, which based on extrawater content in a tissue. They include evaluation and physical examination, by assessing volume and shape of the extremities, tissue impedance control in the kHz-MHz range [3].

It should take into account that tissue fibrosis is emerged before clinical manifestations of lymphedema [4]. Methods of optical imaging, such as optical coherent tomography, multi-photon microscopy etc. have spatial resolution of several microns, that is enough for tissue fine structure control.

Evaluation of changing of tissue structure can be based on methods of image analysis and image recognition. The machine learning allows to provide effective image based diagnosis. We plan to discuss most optimal approaches for image segmentation, informative features extraction and predictive model construction, which is suitable for early lymphedema detection. The work was carried out under partial financial support of the Russian Fund of Basic Research (grant No.17-00-00186, grant No. 18-42-703012.


[1] M.Caruana, Lipedema: A Commonly Misdiagnosed Fat Disorder, Plastic Surgical Nursing, 38 (4), pp. 149-152 (2018).

[2] J. E. Tisaire, E.M. Rodrigo, S.Ribeiro et al., Concept Design of a New Portable Medical Device for Lymphedema Monitoring: A EIT Health ClinMed Summer School Project, Proceedings of the 12th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC Special Session on Designing Future Health Innovations as Needed), pp. 611-620 (2019).

[3] E.S. Qin, M.J. Bowen, W. F. Chen, Diagnostic accuracy of bioimpedance spectroscopy in patients with lymphedema: A retrospective cohort analysis, Journal of Plastic, Reconstructive & Aesthetic Surgery, 71, pp.1041-1050 (2018).

[4] M. Mihara et al., Pathological steps of cancer-related lymphedema: histological changes in the collecting lymphatic vessels after lymphadenectomy, PLoS One. 7 (7), e41126 (2012).

Завантажити оригінал статті: