Leland Stanford Junior University

General Electric Company

University of California

Ge Medical Systems Global Technology Company, LLC

Ge Precision Healthcare LLC

John Mark Pauly

Shreyas S Vasanawala

Michael Lustig

Joseph Yitan Cheng

Greg Zaharchuk

Enhao Gong

Dwight G Nishimura

Tao Zhang

Adam Bruce Kerr

Daniel Mark Spielman

Daniel Blackburn Vigneron

Craig H Meyer

Charles H Cunningham

Steven M Conolly

Priti Balchandani

Graham Arnold Wright

Jean Helen Brittain

Peder E Larson

Feiyu Chen

Juan Manuel Santos

Morteza Mardani Korani

Norbert Joseph Pelc

Albert Macovski

David Leigh Donoho

Jin Hyung Lee

Brian A Hargreaves

Kim Butts Pauly

Suchandrima Banerjee

Rosemary Kim Butts

Pablo Irarrazabal

Douglas C Noll

Karla L Miller

Yuxin Chen

Albert P Chen

Reed Frederick Busse

Krishna S Nayak

Marcus Tedrow Alley

Bob S Hu

John Andrew Derbyshire

Garry Evan Gold

David Henry Gurr

Sanjay Mani

Christopher Michael Sandino

Valentina Taviani

Wenmiao Lu

Frank Ong

Jeffrey Alan Stainsby

Seung Jean Kim

Zhong-Min Hu

Sonal Josan

Kim Butts

Brain A Hargreaves

Kangrong Zhu

John Star-Lack

Patrick L Le Roux

Gerard T Luk Pat

Elizabeth K Cole

Walter F Block

A method for diagnostic imaging with reduced contrast agent dose uses a deep learning network (DLN) [] that has been trained using zero-contrast [] and low-contrast [] images as input to the DLN and full-contrast images [] as reference ground truth images. Prior to training, the images are pre-processed [] to co-register and normalize them. The trained DLN [] is then used to predict a synthesized full-dose contrast agent image [] from acquired zero-dose and low-dose images.

Contrast dose reduction for medical imaging using deep learning

Methods and systems for synthesizing contrast images from a quantitative acquisition are disclosed. An exemplary method includes performing a quantification scan, using a trained deep neural network to synthesize a contrast image from the quantification scan, and outputting the contrast image synthesized by the trained deep neural network. In another exemplary method, an operator can identify a target contrast type for the synthesized contrast image. A trained discriminator and classifier module determines whether the synthesized contrast image is of realistic image quality and whether the synthesized contrast image matches the target contrast type.

System and method for synthesizing magnetic resonance images

A method of reducing radiation dose for radiology imaging modalities and nuclear medicine by using a convolutional network to generate a standard-dose nuclear medicine image from low-dose nuclear medicine image, where the network includes N convolution neural network (CNN) stages, where each stage includes M convolution layers having K×K kernels, where the network further includes an encoder-decoder structure having symmetry concatenate connections between corresponding stages, downsampling using pooling and upsampling using bilinear interpolation between the stages, where the network extracts multi-scale and high-level features from the low-dose image to simulate a high-dose image, and adding concatenate connections to the low-dose image to preserve local information and resolution of the high-dose image, the high-dose image includes a dose reduction factor (DRF) equal to 1 of a radio tracer in a patient, the low-dose PET image includes a DRF of at least 4 of the radio tracer in the patient.

Dose reduction for medical imaging using deep convolutional neural networks

A method is disclosed for generating pixel risk maps for diagnostic image reconstruction. The method produces uncertainty/variance maps by feeding into a trained encoder a short-scan image acquired from a medical imaging scan to generate latent code statistics including the mean &#x3bc;and variance &#x3c3;; selecting random samples based on the latent code statistics, z&#x2dc;N(&#x3bc;,&#x3c3;); feeding the random samples into a trained decoder to generate a pool of reconstructed images; and calculating, for each pixel of the pool of reconstructed images, the pixel mean and variance statistics across the pool of reconstructed images. The risk of each pixel may be calculated using the Stein's Unbiased Risk Estimator on the input density compensated data, that involves calculating the end-to-end divergence of the deep neural network.

Methods for risk map prediction in AI-based MRI reconstruction

A method of reducing radiation dose for radiology imaging modalities and nuclear medicine by using a convolutional network to generate a standard-dose nuclear medicine image from low-dose nuclear medicine image, where the network includes N convolution neural network (CNN) stages, where each stage includes M convolution layers having K&#xd7;K kernels, where the network further includes an encoder-decoder structure having symmetry concatenate connections between corresponding stages, downsampling using pooling and upsampling using bilinear interpolation between the stages, where the network extracts multi-scale and high-level features from the low-dose image to simulate a high-dose image, and adding concatenate connections to the low-dose image to preserve local information and resolution of the high-dose image, the high-dose image includes a dose reduction factor (DRF) equal to 1 of a radio tracer in a patient, the low-dose PET image includes a DRF of at least 4 of the radio tracer in the patient.

A method of magnetic resonance imaging acquires undersampled MRI data and generates by an adversarially trained generative neural network MRI data having higher quality without using any fully-sampled data as a ground truth. The generative neural network is adversarially trained using a discriminative neural network that distinguishes between undersampled MRI training data and candidate undersampled MRI training data produced by applying an MRI measurement function containing an undersampling mask to generated MRI training data produced by the generative neural network from the undersampled MRI training data.

MRI image reconstruction from undersampled data using adversarially trained generative neural network

A method for magnetic resonance imaging (MRI) includes steps of acquiring by an MRI scanner undersampled magnetic-field-gradient-encoded k-space data; performing a self-calibration of a magnetic-field-gradient-encoding point-spread function using a first neural network to estimate systematic waveform errors from the k-space data, and computing the magnetic-field-gradient-encoding point-spread function from the systematic waveform errors; reconstructing an image using a second neural network from the magnetic-field-gradient-encoding point-spread function and the k-space data.

Method for estimating systematic imperfections in medical imaging systems with deep learning

A method for diagnostic imaging with reduced contrast agent dose uses a deep learning network (DLN) [that has been trained using zero-contrast [and low-contrast [] images as input to the DLN and full-contrast images [as reference ground truth images. Prior to training, the images are pre-processed [to co-register and normalize them. The trained DLN [is then used to predict a synthesized full-dose contrast agent image [] from acquired zero-dose and low-dose images.

A method for providing magnetic resonance imaging with dynamic contrast and 4D flow of a volume of an object in a magnetic resonance imaging (MRI) system is provided. Contrast agent is provided to the volume of the object. Magnetic resonance excitation from the MRI system is applied to the volume of the object. The MRI system reads out a subsample of less than 10% of spatially resolved data and velocity encoded data with respect to time. The readout subsample is used to determine both dynamic contrast and 4D flow.

Dynamic contrast enhanced magnetic resonance imaging with flow encoding

A method for diagnostic imaging includes measuring undersampled data y with a diagnostic imaging apparatus; linearly transforming the undersampled data y to obtain an initial image estimate {tilde over (x)}; applying the initial image estimate {tilde over (x)} as input to a generator network to obtain an aliasing artifact-reduced image x&#x306; as output of the generator network, where the aliasing artifact-reduced image x&#x306; is a projection onto a manifold of realistic images of the initial image estimate {tilde over (x)}; and performing an acquisition signal model projection of the aliasing artifact-reduced x&#x306; onto a space of consistent images to obtain a reconstructed image {circumflex over (x)} having suppressed image artifacts.

MRI reconstruction using deep learning, generative adversarial network and acquisition signal model

A system for magnetic resonance imaging an object via a stochastic optimization of a sampling function is provided. The system includes a magnet assembly and a controller. The magnet assembly is operative to acquire MR data from the object. The controller is operative to: acquire a first MR data set using the magnet assembly; select the sampling function from a plurality of sampling function candidates based at least in part on the stochastic optimization; and acquire a second MR data set from the object using the magnet assembly based at least in part on the sampling function.

System and method for magnetic resonance imaging an object via a stochastic optimization of a sampling function

A method for magnetic resonance imaging performs unsupervised training of a deep neural network of an MRI apparatus using a training set of under-sampled MRI scans, where each scan comprises slices of under-sampled, unclassified k-space MRI measurements. The MRI apparatus performs an under-sampled scan to produce under-sampled k-space data, updates the deep neural network with the under-sampled scan, and processes the under-sampled k-space data by the updated deep neural network of the MRI apparatus to reconstruct a final MRI image.

Method for performing magnetic resonance imaging reconstruction with unsupervised deep learning

A method for magnetic resonance imaging acquires multi-channel subsampled k-space data using multiple receiver coils; performs singular-value-decomposition on the multi-channel subsampled k-space data to produce compressed multi-channel k-space data which normalizes the multi-channel subsampled k-space data; applies a first center block of the compressed multi-channel k-space data as input to a first convolutional neural network to produce a first estimated k-space center block that includes estimates of k-space data missing from the first center block; generates an n-th estimated k-space block by repeatedly applying an (n&#x2212;1)-th estimated k-space center block combined with an n-th center block of the compressed multi-channel k-space data as input to an n-th convolutional neural network to produce an n-th estimated k-space center block that includes estimates of k-space data missing from the n-th center block; reconstructs image-space data from the n-th estimated k-space block.

Generalized multi-channel MRI reconstruction using deep neural networks

A method for performing wave-encoded magnetic resonance imaging of an object is provided. The method includes applying one or more wave-encoded magnetic gradients to the object, and acquiring MR signals from the object. The method further includes calibrating a wave point-spread function, and reconstructing an image from the MR signals based at least in part on the calibrated wave point-spread function. Calibration of the wave point-spread function is based at least in part on one or more intermediate images generated from the MR signals.

System and method for performing wave-encoded magnetic resonance imaging of an object

A method of improving diagnostic and functional imaging is provided by obtaining at least two input images of a subject, using a medical imager, where each input image includes a different contrast, generating a plurality of copies of the input images using non-local mean (NLM) filtering, using an appropriately programmed computer, where each input image copy of the subject includes different spatial characteristics, obtaining at least one reference image of the subject, using the medical imager, where the reference image includes imaging characteristics that are different form the input images of the subject, training a deep network model, using data augmentation on the appropriately programmed computer, to adaptively tune model parameters to approximate the reference image from an initial set of the input and reference images, with the goal of outputting an improved quality image of other sets of low SNR low resolution images, for analysis by a physician.

Quality of medical images using multiple-contrast and deep learning

Growing community of inventors

Stanford, CA, United States of America

John Mark Pauly