[Development of an automated processing method to detect coronary motion for coronary magnetic resonance angiography]

On coronary MR angiography (CMRA), cardiac motions worsen the image quality. To improve the image quality, detection of cardiac especially for individual coronary motion is very important. Usually, scan delay and duration were determined manually by the operator. We developed a new evaluation method to calculate static time of individual coronary artery.

At first, coronary cine MRI was taken at the level of about 3 cm below the aortic valve (80 images/R-R). Chronological change of the signals were evaluated with Fourier transformation of each pixel of the images were done. Noise reduction with subtraction process and extraction process were done. To extract higher motion such as coronary arteries, morphological filter process and labeling process were added. Using these imaging processes, individual coronary motion was extracted and individual coronary static time was calculated automatically. We compared the images with ordinary manual method and new automated method in 10 healthy volunteers.

Coronary static times were calculated with our method. Calculated coronary static time was shorter than that of ordinary manual method. And scan time became about 10% longer than that of ordinary method. Image qualities were improved in our method.

Our automated detection method for coronary static time with chronological Fourier transformation has a potential to improve the image quality of CMRA and easy processing.