While high-resolution proton density-weighted magnetic resonance imaging (MRI) of intracranial vessel walls is significant for a precise diagnosis of intracranial artery disease, its long acquisition time is a clinical burden. Compressed sensing MRI is a prospective technology with acceleration factors that could potentially reduce the scan time. However, high acceleration factors result in degraded image quality. Although recent advances in deep-learning-based image restoration algorithms can alleviate this problem, clinical image pairs used in deep learning training typically do not align pixel-wise. Therefore, in this study, two different deep-learning-based denoising algorithms-self-supervised learning and unsupervised learning-are proposed; these algorithms are applicable to clinical datasets that are not aligned pixel-wise. The two approaches are compared quantitatively and qualitatively. Both methods produced promising results in terms of image denoising and visual grading. While the image noise and signal-to-noise ratio of self-supervised learning were superior to those of unsupervised learning, unsupervised learning was preferable over self-supervised learning in terms of radiomic feature reproducibility.

Abstract:We report efficient acousto-optic control of speckle contrast at the output of multimode fibers (MMFs) using a cylindrical piezoelectric transducer (PZT) vibrating in the radial direction. With appropriate packaging of an MMF around the PZT, periodic stretching and subsequent intensity modulation were achieved over the wound fiber to result in timeaveraged smoothing of the output within a short time. It was experimentally confirmed that light passed through the vibrating PZT-fiber assembly maintains the virtually partial coherence irrespective of the guide length and power splitting. Real-time vibration-off/on movies were presented, and their single-frame excerpts were analyzed. Electron. Lett. 15(23), 755-756 (1979). 13. E. G. Rawson, J. W. Goodman, and R. E. Norton, "Analysis and measurement of the modal-noise probability distribution for a step-index optical fiber," Opt. Lett. 5(8), 357-358 (1980). 14. D. U. Kim, S. C. Bae, J. Kim, T.-Y. Kim, C.-S. Park, and K. Oh, "Hard polymer cladding fiber (HPCF) links for high-speed short reach 1×4 passive optical network (PON) based on all-HPCF compatible fused taper power splitter," IEEE Photon. ©2009 Optical Society of America