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

Abstract

The discrimination of tumor-infiltrated tissue from non-tumorous brain tissue during neurosurgical tumor excision is a major challenge in neurosurgery. It is critical to achieve full tumor 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 that enables the classification of different levels of tumor infiltration and non-tumorous tissue. This work investigated two OCT systems with different imaging wavelengths (930 nm/1310 nm) and different resolutions (axial (air): 4.9 μm/16 μm, lateral: 5.2 μm/22 μm) in their ability to identify different levels of tumor infiltration based on freshly excised ex vivo brain samples. A convolutional neural network was used for the classification. For both systems, the neural network could achieve classification accuracies above 91% for discriminating between healthy white matter and highly tumor infiltrated white matter (tumor infiltration >60%) .This work shows that both OCT systems with different optical properties achieve similar results regarding the identification of different stages of brain tumor infiltration.

Dual wavelength analysis and classification of brain tumor tissue with optical coherence tomography

“…The reason could be that the combination of the used imaging wavelengths was not optimal for the classification task and that the different resolution properties of the two OCT systems were not sufficient in displaying significant differences between the two tissue types. The discrimination of white matter on the other hand improved with the multimodal setup compared to the results from the monomodal approach presented in [8].…”

“…Yashin et al first showed that it is difficult to discriminate gray matter from tumor since both have similar optical properties due to the lack of myelin fibers [6]. In the previous work the monomodal classification approach of healthy tissue (gray matter and white matter) and tumor infiltrated white matter a sensitivity of 0.58 ± 0.20 and a specificity of 0.63 ± 0.27 for the SS-OCT and a sensitivity of 0.54 ± 0.19 and a specificity of 0.65 ± 0.17 for the SD-OCT [8]. The multimodal approach shows similar results for the classification of tumor infiltration, when considering all the data.…”

“…The optical properties include the attenuation coefficient, which describes the signal decay with increasing light penetration and the backscattered intensity, which initial intensity of the OCT signal before it decays. Both parameters have shown in the past publications to be meaningful features when it comes to the identification of brain tissue [6,8]. Healthy white matter is characterized by a high attenuation coefficient and a high backscattering intensity that is why the network reacts more to the features at the top of the image.…”

“…Figure 1 shows an overview on how the multimodal dataset for the tissue classification was created. In a first step each OCT image was corrected for focus and roll-off effects as shown in [8]. This helps to reduce intensity differences in an OCT image, which result from different tissue positions within the image.…”

“…Over the last decade several research groups demonstrated, that optical coherence tomography (OCT) could be a possible additional imaging method for the detection of the tumor boundaries [3][4][5][6]. In recent publications we introduced a dataset, which consisted of corresponding OCT images from two OCT Systems [7,8]. Detailed histological labels were transformed from histological images onto corresponding OCT images.…”