CC BY
5The 30th International Conference on Advanced Laser Technologies
ALT'23
B-O-2
Feature analysis of OCT images for the diagnosis of brain glioma
P.V. Aleksandrova1*, K.I. Zaytsev1, P.V. Nikitin2, A.I. Alekseeva3, I.V. Reshetov4, I.N.
Dolganova5
1-Prokhorov General Physics Institute of the Russian Academy of Sciences, Moscow 119991, Russia 2- University of Houston, Houston, Texas 77204, USA 3- Research Institute of Human Morphology, Moscow 117418, Russia 4 - Institute for Cluster Oncology, Sechenov University, Moscow 119991, Russia 5 - Institute of Solid State Physics of the Russian Academy of Sciences, Chernogolovka 142432, Russia
Optical coherence tomography (OCT) is a fast non-invasive method which is used to visualize the internal structure of tissues [1]. OCT has found its application in various clinical fields, including neurosurgery [25]. An intraoperative diagnosis of brain tumors is one of the most urgent and challenging problem of modern neurosurgery [6-8]. Existing methods of the intraoperative neurodiagnosis of tumors are plagued with limited sensitivity or still rather expensive. Thus, a development of novel intraoperative diagnostic methods of brain gliomas, aimed at demarcation of tumor boundaries is one of the most important tasks of medicine, physics, and engineering sciences.
In our research, aimed at the application of OCT for the diagnosis of brain gliomas of different grades, we obtained OCT signals for ex vivo samples of brain glioma of various grades and intact brain tissue. In particular, we proposed a set of features for tissue differentiation, namely, the attenuation coefficient and its variance within the sample, and the local brightness fluctuations in OCT speckle patterns, obtained by means of the wavelet analysis. Then we applied the linear discriminant analysis to compare the advantages and weaknesses of particular features for distinguishing different tissue types. The results of this study confirmed the perspectives of combined attenuation-speckle signal analysis for neurosurgical purposes.
'T "iWflii i?L" i
OCT glioma grade 3
(b)
r 0 0.2
H&E
Is '
glioma 3 /(e)
/
white matter (f)
>É-<
0 0.1 0.2 0.3 0.4 0.5 5 5.5 6
mm"1 P1x 10 5, arb.units
Fig. 2. An example of obtained results: (a) OCT images of glioma grade 3; (b) representative H&E-stained histological image; (c), (d) distribution of the attenuation properties p and a^ for glioma grade 3, cortex and white matter, respectively; (e), (f) distribution of speckle properties Pa and aPa for glioma grade 3, cortex and white matter, respectively.
This work was supported by the Russian Science Foundation (RSF), research project # 19-79-10212.
[1] V. V. Tuchin Tissue Optics: Light Scattering Methods and Instruments for Medical Diagnostics, 3rd ed., SPIE Press, (2015).
[2] F. Fercher, W. Drexler, C. K. Hitzenberger, Optical coherence tomography - principles and applications, T. Lasser, Reports on Progress in Physics, 66 (2), 239, (2003).
[3] S. A. Boppart, Optical coherence tomography: technology and applications for neuroimaging, Psychophysiology, 40 (4), 529-541(2003).
[4] W. Drexler, M. Liu, A. Kumar, et al., Optical coherence tomography today: speed, contrast, and multimodality, J. Biomed. Opt., 19 (7), 071412, (2014).
[5] I.N. Dolganova, P.V. Aleksandrova, P.V. Nikitin, et al, Capability of physically reasonable OCT-based differentiation between intact brain tissues, human brain gliomas of different WHO grades, and glioma model 101.8 from rats, Biomed. Opt. Exp., 11, 6780-6798 (2020).
[6] X. Yu, C. Hu, W. Zhang, et al., Feasibility evaluation of micro-optical coherence tomography ^OCT for rapid brain tumor type and grade discriminations: ^OCT images versus pathology, BMC Medical Imaging, 19, 102, (2019).
[7] J. Wang, Y. Xu, S. A Boppart, Review of optical coherence tomography in oncology, J. Biomed. Opt., 22 (12), 1-23 (2017).
[8] A. Majumdar, N. Allam, W.J. Zabel, et al., Binary dose level classification of tumour microvascular response to radiotherapy using artificial intelligence analysis of optical coherence tomography images, Scientific Reports, 12, 13995, (2022).
0.2
0.l
6.5
