Toshiki Saito BS, RT, Presenter: Nothing to Disclose

Yuki Ichinoseki MS, RT, Abstract Co-Author: Nothing to Disclose

Kota Miyamoto BS, RT, Abstract Co-Author: Nothing to Disclose

Tatsuo Nagasaka RT, Abstract Co-Author: Nothing to Disclose

Yoshio Machida PhD, Abstract Co-Author: Nothing to Disclose

Purposes of this exibit are the following. 1. To review image quality (IQ) evaluation in recently introduced compressed sensing (CS) MRI 2. To recognize the necessity for evaluation aimed at clinical application 3. To learn useful examples related to evaluation of the depictability in MR angiography (MRA) using numerical simulation

CS-MRI and conventional IQ evluation - CS: Sophisticated theory in informatics using data sparsity - CS-MRI: Complicated nonlinear procedure enabling fast scan   Various results depending on an assumed model - IQ indexes: MTF, NPS or "P-SNR, SSIM" in informatics   Appropliate evaluation for clinical images? IQ evaluation aimed at clinical applications in CS-MRI - Our concern for evaluation in clinical images such as low contrast in T2WI and thin blood vessels in MRA - Direct evaluation is useful under complicated circumstances Analysis of the depictability of thin blood vessels: Numerical simulation - Simple numerical model mimicking cerebral arteries was used - MRA images were obtained using various MRA and CS sparsity conditions   Bright / Black Blood, and subtracted MRA   CS sparsity in wavelet space or image domain - Elucidation of CS-MRI characteristics   Blood vessel depictability in MIP (or minIP) image   Semi-quantitative evaluation with signal intensity of vessels

http://abstract.rsna.org/uploads/2014/14009001/14009001_ifrh.pdf

Saito, T, Ichinoseki, Y, Miyamoto, K, Nagasaka, T, Machida, Y, Evaluating Image Quality of Compressed Sensing MRI Aiming at Clinical Applications.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14009001.html