US 11,789,104 B2
Deep learning techniques for suppressing artefacts in magnetic resonance images
Carole Lazarus, Paris (FR); Prantik Kundu, Branford, CT (US); Sunli Tang, New York, NY (US); Seyed Sadegh Mohseni Salehi, Bloomfield, NJ (US); Michal Sofka, Princeton, NJ (US); Jo Schlemper, Long Island City, NY (US); Hadrien A. Dyvorne, New York, NY (US); Rafael O'Halloran, Guilford, CT (US); Laura Sacolick, Guilford, CT (US); Michael Stephen Poole, Guilford, CT (US); and Jonathan M. Rothberg, Miami Beach, FL (US)
Assigned to Hyperfine Operations, Inc., Guilford, CT (US)
Filed by Hyperfine Operations, Inc., Guilford, CT (US)
Filed on Aug. 15, 2019, as Appl. No. 16/541,511.
Claims priority of provisional application 62/820,119, filed on Mar. 18, 2019.
Claims priority of provisional application 62/764,742, filed on Aug. 15, 2018.
Prior Publication US 2020/0058106 A1, Feb. 20, 2020
Int. Cl. G06V 10/30 (2022.01); G01R 33/56 (2006.01); G06T 5/00 (2006.01); G06N 3/045 (2023.01); G06V 10/764 (2022.01); G06V 10/772 (2022.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01); G06V 10/44 (2022.01)
CPC G01R 33/5608 (2013.01) [G06N 3/045 (2023.01); G06T 5/002 (2013.01); G06V 10/30 (2022.01); G06V 10/454 (2022.01); G06V 10/764 (2022.01); G06V 10/772 (2022.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01); G06T 2207/10088 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method, comprising:
obtaining input magnetic resonance (MR) data using at least one radio-frequency (RF) coil of a magnetic resonance imaging (MRI) system; and
generating an MR image from the input MR data at least in part by using a neural network model to suppress RF interference in the input MR data, wherein the generating comprises:
performing a reconstruction step by generating an image in an image domain from the input MR data; and
using the neural network model to suppress the RF interference in the input MR data by processing the input MR data before the reconstruction step in a domain other than the image domain,
wherein the neural network model comprises:
a first neural network portion configured to suppress the RF interference in the input MR data in the domain other than the image domain before the reconstruction step, the first neural network portion comprising one or more convolutional layers; and
a second neural network portion configured to suppress noise in the input MR data in the domain other than the image domain before the reconstruction step, the second neural network portion comprising one or more convolutional layers.