SimNICT

SimNICT is the first dataset for training universal non-ideal measurement CT (NICT) enhancement models.

The dataset comprises over 10.9 million NICT-ICT image pairs, including low dose CT (LDCT), sparse view CT (SVCT), and limited angle CT (LACT), under varying defect degrees across whole-body regions.

Paper: Imaging foundation model for universal enhancement of non-ideal measurement CT