Accuracy Confirmation of the Compress-Sensing Technique in the TOF MRA Test using Flow Phantom

This study evaluated the effects of applying the compress-sensing technique and the increase of the CS factor on the quality of the image in the test using the time-of-flight sequence. A 3D TOF MRA test was performed while maintaining a constant flow rate (2.0 ml/sec) by connecting an auto-injector to a self-fabricated flow phantom. Images were obtained by applying the 3D TOF sequence without CS and with CS to evaluate the difference in image quality with or without application of the CS technique. Moreover, in order to analyze the quality of images according to the CS factor, images were obtained while increasing the CS factor from 1.2 to 1.8 by 0.2. The examination time was 44 seconds when the CS technique was not used. When the CS technique was applied, the examination time decreased to 30, 27, 25, and 23 with the CS factor of 1.2, 1.4, 1.6, and 1.8, respectively. On the other hand, the application of the CS technique did not change the SNR or the CNR of the image significantly (p > 0.05). Moreover, the SNR and the CNR of the image were not significantly affected by the changes in the CS factor (p > 0.05). However, as the CS factor increased, the similarity and precision decreased more than in the images where the CS technique was not used. The application of the CS technique to the TOF MRA test can reduce the examination time drastically without changing SNR and CNR. However, we found that, as the CS factor increased, the similarity and precision decreased more than in the test without the CS technique. Therefore, it is necessary to continuously verify the effectiveness of the CS technique and study it more.