Enhanced Optical Coherence Tomography (OCT) of Prostate Nerves Through Integrated Image-Processing Techniques

Cavernous nerves, located along the prostate gland's surface, are integral to erectile functionality. These nerves are at risk of injury during the surgical removal of a cancerous prostate gland. This research applies a suite of image processing algorithms—segmentation, denoising, and edge detection—to time-domain optical coherence tomography (OCT) images of prostates from different datasets of to improve the detection of cavernous nerves. Initially, the prostate OCT images are segmented to isolate the cavernous nerves from the adjacent glandular tissue. This is followed by the application of a locally adaptive denoising process using a dual-tree complex wavelet transform, aimed at reducing speckle noise. Subsequently, edge detection techniques are employed to enhance the imaging depth of the prostate gland. The combined application of these image processing techniques significantly improves the signal-to-noise ratio and imaging depth, enabling the automated identification of cavernous nerves. This enhanced imaging capability is crucial for supporting nerve-sparing approaches in laparoscopic and robotic prostate cancer surgeries.