Fast magnetic resonance reconstruction method and system fused with depth sensitivity estimation

By constructing a convolution reconstruction network that integrates deep neural networks, the problem of low efficiency and accuracy of magnetic resonance reconstruction in the prior art is solved, and more efficient and more accurate magnetic resonance image reconstruction is achieved.

CN119672144AActive Publication Date: 2025-03-21SHENZHEN ANKE HIGH TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411646908.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-18
Publication Date
2025-03-21
Estimated Expiration
2044-11-18

AI Technical Summary

Technical Problem

In the prior art, due to the performance constraints of the sensitivity estimation calculation method, the magnetic resonance reconstruction efficiency and accuracy are not high.

Method used

The fast magnetic resonance reconstruction method of fused deep neural network is adopted to build a convolutional reconstruction network through coil sensitivity estimation network, channel synthesis module and cascade reconstruction network, realize multi-task joint training, and improve reconstruction accuracy and robustness.

Benefits of technology

It effectively improves the accuracy and robustness of magnetic resonance reconstruction, shortens scanning time, improves reconstruction efficiency, and accelerates the convergence speed of the model through multi-task learning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119672144A_ABST
    Figure CN119672144A_ABST
Patent Text Reader

Abstract

The invention provides a fast magnetic resonance reconstruction method and system fused with depth sensitivity estimation, and particularly relates to the technical field of magnetic resonance imaging and image reconstruction, and the method comprises the steps: collecting multi-channel under-sampling K space data; inverse Fourier transform is carried out on the multi-channel under-sampling K space data, and a multi-channel under-sampling magnetic resonance image is obtained; and inputting the multi-channel undersampled magnetic resonance image into a trained convolutional reconstruction network for reconstruction to obtain a reconstructed magnetic resonance image, the convolutional reconstruction network being constructed based on a deep neural network and being obtained by fusing deep sensitivity estimation training. According to the scheme, the undersampling mode is adopted to collect the magnetic resonance K space data as the data set, so that the scanning time can be shortened, the reconstruction efficiency can be improved, and the reconstruction precision and robustness of the system can be effectively improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Multi-channel magnetic resonance image reconstruction method based on deep learning

    CN110378980A

  • Fast magnetic resonance image reconstruction method based on ultrafast Fourier convolution

    CN115908614A

  • Magnetic resonance image reconstruction device and magnetic resonance image reconstruction method

    CN118071849A