Deep Learning-Based Algorithm for Electromagnetic Interference Noise Removal in Photoacoustic Endoscopic Image Processing

Abstract

In this study, we developed deep-learning-based image processing algorithms to remove the electromagnetic interference (EMI) noise included in optical-resolution (OR) photoacoustic endoscopy (PAE) images, and we have witnessed that thereby EMI noise can be significantly removed from those images. Although we do not emphasize this point, engineering problems related to EMI noise form an important and fundamental subject area in electronics since EMI noise frequently intervenes between a sensor and an amplifier. To the best of our knowledge, this paper is the first to deal with the question of removing EMI noise from PAT images by using deep learning techniques.