Siemens Healthcare Gmbh

Siemens Healthineers Ag

X-cutag Therapeutics, Inc.

Xiao Chen

Mariappan S Nadar

Boris Mailhe

Benjamin L Odry

Qiu Wang

Simon Arberet

Peter Speier

Hasan Ertan Cetingul

Pascal Ceccaldi

Sandro Braun

Marcel Dominik Nickel

Mathias Nittka

Xiaoguang Lu

Qiaoying Huang

Silvia Bettina Arroyo Camejo

Guillaume Daval Frerot

Bogdan Georgescu

Puneet Sharma

Shaohua Kevin Zhou

Yefeng Zheng

Changfu Cheng

Yi Guo

Hui Joyce Li

Shuhao Wen

Christian Alexander Braun

Gao Shang

Christopher Cline

Julia Traechtler

Yevgen Matviychuk

LuoLuo Liu

For detecting motion in MR imaging, a regression model, such as a convolutional neural network, is machine trained. To generalize to MR imagers, MR contrasts, or other differences in MR image generation, the regression model is trained adversarially. The discriminator for adversarial training discriminates between classes of the variation source (e.g., type of MR imager or type of contrast) based on values of features learned in the regression model for detecting motion. By adversarial training, the regression model learns features that are less susceptible or invariant to variation in image source.

Contrast and/or system independent motion detection for magnetic resonance imaging

For machine training and application of a trained complex-valued machine learning model, an activation function of the machine learning model, such as a neural network, includes a learnable parameter that is complex or defined in a complex domain with two dimensions, such as real and imaginary or magnitude and phase dimensions. The complex learnable parameter is trained for any of various applications, such as MR fingerprinting, other medical imaging, or non-medical uses.

Complex-valued neural network with learnable non-linearities in medical imaging

For k-space trajectory infidelity correction, a model is machine trained to correct k-space measurements in k-space. K-space trajectory infidelity correction uses deep learning. Trajectory infidelity is corrected from a k-space point of view. Since the image artifacts arise from k-space acquisition distortion, a machine learning model is trained to correct in k-space, either changing values of k-space measurements or estimating the trajectory shifts in k-space.

K-space trajectory infidelity correction in magnetic resonance imaging

Machine training a network for and use of the machine-trained network are provided for tissue parameter estimation for a magnetic scanner using magnetic resonance fingerprinting. The machine-trained network is trained to both reconstruct a fingerprint image or fingerprint and to estimate values for multiple tissue parameters in magnetic resonance fingerprinting. The reconstruction of the fingerprint image or fingerprint may reduce noise, such as aliasing, allowing for more accurate estimation of the values of the multiple tissue parameters from the under sampled magnetic resonance fingerprinting information.

Magnetic resonance fingerprinting image reconstruction and tissue parameter estimation

For determination of motion artifact in MR imaging, motion of the patient in three dimensions is used with a measurement k-space line order based on one or more actual imaging sequences to generate training data. The MR scan of the ground truth three-dimensional (3D) representation subjected to 3D motion is simulated using the realistic line order. The difference between the resulting reconstructed 3D representation and the ground truth 3D representation is used in machine-based deep learning to train a network to predict motion artifact or level given an input 3D representation from a scan of a patient. The architecture of the network may be defined to deal with anisotropic data from the MR scan.

Motion determination for volumetric magnetic resonance imaging using a deep machine-learning model

Systems and methods are provided for iterative reconstruction of a magnetic resonance image using magnetic resonance fingerprinting. An image series is estimated according to the following four steps: a gradient step to improve data consistency, fingerprint matching, spatial regularization, and a merging step. The fingerprint matching and spatial regularization steps are performed in parallel.

Magnetic resonance fingerprinting based on bloch manifold and spatial regularizations in parallel

For classifying magnetic resonance image quality or training to classify magnetic resonance image quality, deep learning is used to learn features distinguishing between corrupt images base on simulation and measured similarity. The deep learning uses synthetic data without quality annotation, allowing a large set of training data. The deep-learned features are then used as input features for training a classifier using training data annotated with ground truth quality. A smaller training data set may be needed to train the classifier due to the use of features learned without the quality annotation.

Magnetic resonance imaging quality classification based on deep machine-learning to account for less training data

Various approaches provide improved segmentation from raw data. Training samples are generated by medical imaging simulation from digital phantoms. These training samples provide raw measurements, which are used to learn to segment. The segmentation task is the focus, so image reconstruction loss is not used. Instead, an attention network is used to focus the training and trained network on segmentation. Recurrent segmentation from the raw measurements is used to refine the segmented output. These approaches may be used alone or in combination, providing for segmentation from raw measurements with less influence of noise or artifacts resulting from a focus on reconstruction.

Medical image segmentation from raw data using a deep attention neural network

For determining a configuration for a MR scanner using MRF, MRF is optimized using a model of acquisition and reconstruction together. The effects of under sampling and reconstruction are included in the model being optimized with the sequence. A spatial distribution of response is used in the MRF optimization. An added transformation may be included to convert the fingerprint into a domain for noise removal and/or more direct parameter estimation without the dictionary. Artificial intelligence may be used to alter and/or select the type of optimization or optimization settings and/or to optimize.

Magnetic resonance fingerprinting optimization in magnetic resonance imaging

Systems and methods are provided for iterative reconstruction of a magnetic resonance image using Magnetic Resonance Fingerprinting (MRF). An image series is estimated according to the following three steps: a gradient step to improve data consistency, fingerprint matching, and a spatial regularization. Singular Value Decomposition (SVD) compression may be used along the time dimension to accelerate both the matching and the spatial regularization that operates in the compressed domain as well as to enforce low-rank regularization.

Low rank and spatial regularization model for magnetic resonance fingerprinting

For artifact reduction in a magnetic resonance imaging system, deep learning trains an image-to-image neural network to generate an image with reduced artifact from input, artifacted MR data. For application, the image-to-image network may be applied in real time with a lower computational burden than typical post-processing methods. To handle a range of different imaging situations, the image-to-image network may (a) use an auxiliary map as an input with the MR data from the patient, (b) use sequence metadata as a controller of the encoder of the image-to-image network, and/or (c) be trained to generate contrast invariant features in the encoder using a discriminator that receives encoder features.

Artifact reduction by image-to-image network in magnetic resonance imaging

A system and method including receiving magnetic resonance (MR) imaging data from a first MR scanner device, the MR imaging data including data for a plurality of MR scans of different structural or anatomical regions; generating, based on the MR imaging data, normalized reference data including statistical information for each MR scan; learning a transformation, based on the normalized reference data, to correlate a set of input MR imaging data to the normalized reference data; and storing a record of the transformed imaging data.

System and method for normalized reference database for MR images via autoencoders

Systems and methods are provided for generating segmented output from input regardless of the resolution of the input. A single trained network is used to provide segmentation for an input regardless of a resolution of the input. The network is recursively trained to learn over large variations in the input data including variations in resolution. During training, the network refines its prediction iteratively in order to produce a fast and accurate segmentation that is robust across resolution differences that are produced by MR protocol variations.

Deep reinforcement learning for recursive segmentation

A system and method includes generation of one or more motion-corrupted images based on each of a plurality of reference images, and training of a regression network to determine a motion score, where training of the regression network includes input of a generated motion-corrupted image to the regression network, reception of a first motion score output by the regression network in response to the input image, and determination of a loss by comparison of the first motion score to a target motion score, the target motion score calculated based on the input motion-corrupted image and a reference image based on which the motion-corrupted image was generated.

Image quality assessment system and method

Systems and methods are provided for correcting motion artifacts in magnetic resonance images. An image-to-image neural network is used to generate motion corrected magnetic resonance data given motion corrupted magnetic resonance data. The image-to-image neural network is coupled within an adversarial network to help refine the generated magnetic resonance data. The adversarial network includes a generator network (the image-to-image neural network) and a discriminator network. The generator network is trained to minimize a loss function based on a Wasserstein distance when generating MR data. The discriminator network is trained to differentiate the motion corrected MR data from motion artifact free MR data.

Growing community of inventors

Princeton, NJ, United States of America

Xiao Chen