Portable MRI scanners have the advantages of maximizing clinical availability in remote environments. We have recently developed a portable, elbow scanner installed in a standard-size car. This system allows us to detect early symptoms of baseball elbow in remote places, but it often suffers from the low signal-to-noise ratio in noisy, outdoor environments. Here we proposed a deep-learning based approach, a denoising convolutional neural network with transfer learning, for denoising images of the potable scanner. We verified that the proposed denoising technique improved the quality of noisy images and increased the clinical feasibility of the portable scanner.
This abstract and the presentation materials are available to members only; a login is required.