Pixel classification method in optical coherence tomography for tumor segmentation and its complementary usage with OCT microangiography

Abstract

A novel machine-learning method to distinguish between tumor and normal tissue in optical coherence tomography (OCT) has been developed. Pre-clinical murine ear model implanted with mouse colon carcinoma CT-26 was used. Structural-image-based feature sets were defined for each pixel and machine learning classifiers were trained using "ground truth" OCT images manually segmented by comparison with histology. The accuracy of the OCT tumor segmentation method was then quantified by comparing with fluorescence imaging of tumors expressing genetically encoded fluorescent protein KillerRed that clearly delineates tumor borders. Because the resultant 3D tumor/normal structural maps are inherently co-registered with OCT derived maps of tissue microvasculature, the latter can be color coded as belonging to either tumor or normal tissue. Applications to radiomics-based multimodal OCT analysis are envisioned.

Method forin vivoassessment of cancer tissue inhomogeneity and accurate histology-like morphological segmentation based on Optical Coherence Elastography

“…can be used (see, e.g. 6 ). Multimodal OCT has also proven to be very useful for in vivo evaluation of fairly rapid and pronounced posttherapeutic changes in tumors.…”

“…Multimodal OCT has also proven to be very useful for in vivo evaluation of fairly rapid and pronounced posttherapeutic changes in tumors. For example, one can mention the OCT-based angiographic observation of blood-circulation blockages in tumors and in peri-tumorous regions for accurate prediction of the outcome of vasculature-targeted photodynamic therapy (PDT) during the first 24 hours post-PDT [6][7][8] . However, unlike the fairly easily observed perturbation in the microcirculation of blood, the assessment of the histological tissue structure lacks precise, noninvasive methods.…”

Histological validation of in vivo assessment of cancer tissue inhomogeneity and automated morphological segmentation enabled by Optical Coherence Elastography

“…The method eliminates the stage of special long-term and expensive staining 56 . Other studies demonstrate the prospect of using a specially trained neural network to generate a morphological conclusion 5,55 , which makes it possible not to have to resort to the help of a qualified histopathologist and to accelerate the pace of the processing of histological material. In addition, our research team is investigating the possibilities of a cross-polarization OCT method in determining the morphology and boundaries of brain tumors 4 .…”

“…can be used (see, e.g. 5 ). Multimodal OCT has also proven to be very useful for in vivo evaluation of fairly rapid and pronounced post-therapeutic changes in tumors.…”

Optical coherence tomography for precision brain imaging, neurosurgical guidance and minimally invasive theranostics

“…The substantial contribution of this research is the performance of diagnostic analysis to derive an attenuation threshold to distinguish tissues with high sensitivity and specificity. For the identification of tumor margins, some algorithms assist the diagnosis of tumorous tissue, such as pixel classification-based method (84), and attenuation coefficient-based method (85). Machine learning method has been used in the classification of skin tumors with OCT images (86), it has potential application on brain tumor imaging.…”