Systems and methods for joint reconstruction and segmentation of organs from magnetic resonance imaging data
View Patent ↗Systems and methods for joint reconstruction and segmentation of organs from magnetic resonance imaging (MRI) data are provided. Sparse MRI data is received at a computer system, which jointly processes the MRI data using a plurality of reconstruction and segmentation processes. The MRI data is processed using a joint reconstruction and segmentation process to identify an organ from the MRI data. Additionally, the MRI data is processed using a channel-wise attention network to perform static reconstruction of the organ from the MRI data. Further, the MRI data can is processed using a motion-guided network to perform dynamic reconstruction of the organ from the MRI data. The joint processing allows for rapid static and dynamic reconstruction and segmentation of organs from sparse MRI data, with particular advantage in clinical settings.
1. A method for joint reconstruction and segmentation of organs from magnetic resonance imaging (MRI) data, comprising the steps of:
receiving MRI data at a computer system;
processing the MRI data using a joint reconstruction and segmentation process to identify an organ from the MRI data;
processing the MRI data using a channel-wise attention network to perform static reconstruction of the organ from the MRI data; and
processing the MRI data using a motion-guided network to perform dynamic reconstruction of the organ from the MRI data.
2. The method of claim 1 , wherein said processing steps are performed jointly by the computer system.
3. The method of claim 1 , wherein the step of processing the MRI data using the joint reconstruction and segmentation process further comprises processing the MRI data using a first neural network to generate a reconstructed image from the MRI data.
4. The method of claim 3 , wherein the step of processing the MRI data using the joint reconstruction and segmentation process further comprises processing the reconstructed image using a second neural network to identify contours of an organ in the reconstructed image.
5. The method of claim 1 , wherein the step of processing the MRI data using the channel-wise attention network further comprises processing the MRI data using an undersampled mask and undersampled k-space data by the channel-wise attention network.
6. The method of claim 5 , wherein the channel-wise attention network comprises a channel attention layer and a data consistency layer, the data consistency layer cascaded with the channel attention layer and processing outputs of the channel attention layer.
7. The method of claim 1 , wherein the step of processing the MRI data using the motion-guided network comprises processing the MRI data using a recurrent process to dynamically reconstruct the organ in the presence of motion.
8. The method of claim 1 , wherein the step of processing the MRI data using the motion-guided network comprises processing the MRI data using a differentiable network.
9. The method of claim 8 , wherein the differentiable network includes a first layer comprising a dynamic reconstruction network, a second layer comprising a plurality of motion estimation components, and a third layer comprising a motion compensation component.
10. The method of claim 9 , further comprising backpropagating gradients from the motion compensation component to the dynamic reconstruction network and the plurality of motion estimation components to improve training of the differentiable network.
11. A system for joint reconstruction and segmentation of organs from magnetic resonance imaging (MRI) data, comprising:
a memory storing MRI data; and
a processor in communication with the memory, the processor programmed to:
process the MRI data using a joint reconstruction and segmentation process to identify an organ from the MRI data;
process the MRI data using a channel-wise attention network to perform static reconstruction of the organ from the MRI data; and
process the MRI data using a motion-guided network to perform dynamic reconstruction of the organ from the MRI data.
12. The system of claim 11 , wherein processor processes the MRI data using a first neural network to generate a reconstructed image from the MRI data.
13. The system of claim 12 , wherein the processor processes the reconstructed image using a second neural network to identify contours of an organ in the reconstructed image.
14. The system of claim 11 , wherein the processor processes the MRI data using an undersampled mask and undersampled k-space data by the channel-wise attention network.
15. The system of claim 14 , wherein the channel-wise attention network comprises a channel attention layer and a data consistency layer, the data consistency layer cascaded with the channel attention layer and processing outputs of the channel attention layer.
16. The system of claim 11 , wherein the processor processes the MRI data using a recurrent process to dynamically reconstruct the organ in the presence of motion.
17. The system of claim 11 , wherein the processor processes the MRI data using a differentiable network.
18. The system of claim 17 , wherein the differentiable network includes a first layer comprising a dynamic reconstruction network, a second layer comprising a plurality of motion estimation components, and a third layer comprising a motion compensation component.
19. The system of claim 18 , wherein the differentiable network backpropagates gradients from the motion compensation component to the dynamic reconstruction network and the plurality of motion estimation components to improve training of the differentiable network.
20. The system of claim 11 , wherein the processor is in communication with or forms part of a magnetic resonance imaging (MRI) scanner.
21. A method for reconstruction of organs from magnetic resonance imaging (MRI) data, comprising the steps of:
receiving MRI data at a computer system;
processing the MRI data using a channel-wise attention network to perform static reconstruction of the organ from the MRI data; and
displaying the reconstructed organ.
22. The method of claim 21 , wherein the step of processing the MRI data using the channel-wise attention network further comprises processing the MRI data using an undersampled mask and undersampled k-space data by the channel-wise attention network.
23. The method of claim 22 , wherein the channel-wise attention network comprises a channel attention layer and a data consistency layer, the data consistency layer cascaded with the channel attention layer and processing outputs of the channel attention layer.
24. A method for reconstruction of organs from magnetic resonance imaging (MRI) data, comprising the steps of:
receiving MRI data at a computer system;
processing the MRI data using a motion-guided network to perform dynamic reconstruction of the organ from the MRI data; and
displaying the reconstructed organ,
wherein the step of processing the MRI data using the motion-guided network comprises processing the MRI data using a differentiable network.
25. The method of claim 24 , wherein the step of processing the MRI data using the motion-guided network comprises processing the MRI data using a recurrent process to dynamically reconstruct the organ in the presence of motion.
26. The method of claim 24 , wherein the differentiable network includes a first layer comprising a dynamic reconstruction network, a second layer comprising a plurality of motion estimation components, and a third layer comprising a motion compensation component.
27. The method of claim 24 , further comprising backpropagating gradients from the motion compensation component to the dynamic reconstruction network and the plurality of motion estimation components to improve training of the differentiable network.