US Patent Application 18202592. Method for improving high frequency image features and details in deep learning MRI reconstructions simplified abstract

From WikiPatents
Jump to navigation Jump to search

Method for improving high frequency image features and details in deep learning MRI reconstructions

Organization Name

The Board of Trustees of the Leland Stanford Junior University

Inventor(s)

Shreyas S. Vasanawala of Stangord CA (US)

Kanghyun Ryu of Palo Alto CA (US)

Cagan Alkan of Redwood City CA (US)

Method for improving high frequency image features and details in deep learning MRI reconstructions - A simplified explanation of the abstract

This abstract first appeared for US patent application 18202592 titled 'Method for improving high frequency image features and details in deep learning MRI reconstructions

Simplified Explanation

The patent application describes a method for improving magnetic resonance imaging (MRI) using a combination of under-sampled k-space measurements, neural networks, and convex optimization techniques.

  • The method involves acquiring under-sampled k-space measurements from an MRI apparatus using multiple receiver coils.
  • An MRI image is then reconstructed from the under-sampled k-space measurements and coil sensitivity maps using an unrolled neural network.
  • Reconstructed multi-coil k-space data is generated from the MRI image by multiplying it with the coil sensitivity maps and performing a Fourier transform.
  • A k-space null-space convolutional kernel is estimated from fully-sampled k-space measurements in autocalibration signal lines of the under-sampled k-space measurements.
  • A convex optimization problem is solved to produce refined k-space data from the k-space null-space kernel, the under-sampled k-space measurements, and the reconstructed multi-coil k-space data.
  • Finally, a refined MRI image is produced from the refined k-space data by performing an inverse Fourier transform followed by a coil combination using the coil sensitivity maps.


Original Abstract Submitted

A method for magnetic resonance imaging (MRI) includes acquiring under-sampled k-space measurements from an MRI apparatus using multiple receiver coils; reconstructing an MRI image from the under-sampled k-space measurements and coil sensitivity maps using an unrolled neural network; generating reconstructed multi-coil k-space data from the MRI image by multiplying the MRI image by the coil sensitivity maps followed by performing a Fourier transform; estimating a k-space null-space convolutional kernel from fully-sampled k-space measurements in autocalibration signal lines of the under-sampled k-space measurements; solving a convex optimization problem to produce refined k-space data from the k-space null-space kernel, the under-sampled k-space measurements, and the reconstructed multi-coil k-space data; and producing a refined MRI image from the refined k-space data by performing an inverse Fourier transform followed by a coil combination using the coil sensitivity maps.

Retrieved from "http://wikipatents.org/index.php?title=US_Patent_Application_18202592._Method_for_improving_high_frequency_image_features_and_details_in_deep_learning_MRI_reconstructions_simplified_abstract&oldid=15122"