Won Chang

TL;DR: Wang et al. as mentioned in this paper proposed a novel framelet-based denoising algorithm using wavelet residual network which synergistically combines the expressive power of deep learning and the performance guarantee from the framelet based denoizing algorithms.

TL;DR: This is the first study to develop delta-radiomics signatures for rectal cancer and they successfully predicted the outcomes and were independent prognostic factors.

TL;DR: The DLA trained with 50% simulated radiation dose showed the best overall image quality and DLAs achieved less noise than FBP and ADMIRE in LD CT images, but did not maintain spatial resolution.

TL;DR: When read by experienced radiologists, MR can provide reliable diagnostic performance in assessing EMVI for patients with rectal cancer, resulting in better prediction of OS than with pEMVI, with substantial interobserver agreement.

TL;DR: Low monoenergetic spectral CT images (50 keV) can provide better focal liver lesion conspicuity than hybrid iterative reconstruction image of standard-dose CT in nonobese patients while using lower radiation and contrast media doses.