US 12,260,549 B2
Automated deep correction of MRI phase-error
Albert Hsiao, San Diego, CA (US); Evan Masutani, La Jolla, CA (US); and Sophie You, La Jolla, CA (US)
Assigned to THE REGENTS OF THE UNIVERSITY OF CALIFORNIA, Oakland, CA (US)
Filed by THE REGENTS OF THE UNIVERSITY OF CALIFORNIA, Oakland, CA (US)
Filed on Feb. 15, 2022, as Appl. No. 17/672,613.
Claims priority of provisional application 63/149,571, filed on Feb. 15, 2021.
Prior Publication US 2022/0261991 A1, Aug. 18, 2022
Int. Cl. G06T 7/00 (2017.01); G06N 3/08 (2023.01); G06T 7/20 (2017.01); G06V 10/82 (2022.01)
CPC G06T 7/0012 (2013.01) [G06N 3/08 (2013.01); G06T 7/20 (2013.01); G06V 10/82 (2022.01); G06T 2200/04 (2013.01); G06T 2207/10088 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30104 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method for automated correction of phase error in magnetic resonance imaging (MRI)-based flow evaluation, comprising:
receiving in a computer processor configured for executing a trained convolutional neural network (CNN) image data comprising flow velocity data in three directions and magnitude data collected from a region of interest over a scan period from magnetic resonance imaging instrumentation, wherein the region of interest includes static tissue and vessels;
processing the image data with the trained CNN to generate output channels comprising pixelwise inferred corrections corresponding to dimensions of the flow velocity data;
smoothing the pixelwise inferred corrections using a regression algorithm to generate smoothed corrections; and
adding the smoothed corrections to the image data to generate corrected flow data, wherein the corrected flow data is used for flow visualization and quantization.