AUTHOR=Strenge Paul , Lange Birgit , Draxinger Wolfgang , Grill Christin , Danicke Veit , Theisen-Kunde Dirk , Hagel Christian , Spahr-Hess Sonja , Bonsanto Matteo M. , Handels Heinz , Huber Robert , Brinkmann Ralf 

TITLE=Differentiation of different stages of brain tumor infiltration using optical coherence tomography: Comparison of two systems and histology

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.896060

DOI=10.3389/fonc.2022.896060

ISSN=2234-943X

ABSTRACT=Discrimination of tumour infiltrated tissue from non-tumourous brain tissue during the neurosurgical tumour excision is a major challenge in neurosurgery. It is critical to  achieve full tumour removal, since it directly correlates with the survival rate of the patient. Optical coherence tomography (OCT) might be an additional imaging method in the field of neurosurgery, which enables the classification of different levels of tumour infiltration and non-tumourous tissue. This work investigates two OCT systems with different imaging wavelength (930 nm/ 1310 nm) and different resolution (axial(air): 4.9 µm/ 16 µm , lateral: 5.2 µm/ 22 µm) in their ability to identify different levels of tumour infiltration based on freshly excised ex vivo brain samples. For the classification a convolutional neural network was used. For both systems the neural network was able to achieve classification accuracies above 91 % for the discrimination of healthy white matter and highly tumour infiltrated white matter (tumour infiltration >60%). This work shows, that both OCT systems with different optical properties achieve similar results when it comes to the identification of different stages of brain tumour infiltration.