Core needle biopsy guidance using deep-learning assisted OCT imaging (Conference Presentation)

This study presents a portable optical coherence tomography imaging system that provides intraprocedural feedback on the precise selection of the biopsy location. It uses a hand-held optical imaging probe supported by imaging processing pipeline based on deep learning neural networks. The preliminary evaluation of the integrated system was performed on an animal model of cancer, where the collected dataset was used to develop the deep learning-based tumor segmentation software. A patient study for technology clinical validation is on-going at MD Anderson Cancer Center and the preliminary results will be presented.