Few-view CT image reconstruction system
A system for few-view computed tomography (CT) image reconstruction is described. The system includes a preprocessing module, a first generator network, and a discriminator network. The preprocessing module is configured to apply a ramp filter to an input sinogram to yield a filtered sinogram. The first generator network is configured to receive the filtered sinogram, to learn a filtered back-projection operation and to provide a first reconstructed image as output. The first reconstructed image corresponds to the input sinogram. The discriminator network is configured to determine whether a received image corresponds to the first reconstructed image or a corresponding ground truth image. The generator network and the discriminator network correspond to a Wasserstein generative adversarial network (WGAN). The WGAN is optimized using an objective function based, at least in part, on a Wasserstein distance and based, at least in part, on a gradient penalty.
1. A few-view computed tomography (CT) image reconstruction system, the system comprising:
a preprocessing module configured to apply a ramp filter to an input sinogram to yield a filtered sinogram;
a first generator network configured to receive the filtered sinogram, to learn a filtered back-projection operation and to provide a first reconstructed image as an output, the first reconstructed image corresponding to the input sinogram; and
a discriminator network configured to determine whether a received image corresponds to the first reconstructed image or a corresponding ground truth image, the first generator network and the discriminator network corresponding to a Wasserstein generative adversarial network (WGAN), the WGAN optimized using an objective function based, at least in part, on a Wasserstein distance and based, at least in part, on a gradient penalty.
2. The system of claim 1 , further comprising a second generator network configured to receive a concatenation of the first reconstructed image and a filtered back-projection of the input sinogram and to provide a second reconstructed image, the discriminator network further configured to determine whether the received image corresponds to the second reconstructed image.
3. The system of claim 2 , wherein the second generator network corresponds to a refinement portion.
4. The system of claim 2 , further comprising a filtered back projection module configured to receive the input sinogram and to provide the filtered back-projection of the input sinogram.
5. The system of claim 1 , wherein the first generator network is configured to learn the filtered back-projection operation in a point-wise manner.
6. The system of claim 1 , wherein the first generator network comprises a filtration portion, a back-projection portion, and a refinement portion.
7. The system of claim 1 , wherein the WGAN is trained, initially, using image data from an image database comprising a plurality of images.
8. The system of claim 1 , wherein the first generator network is configured to reconstruct the first reconstructed image using O(C×N×N,) parameters, where N is a dimension of the first reconstructed image, N, is a number of projections, and C is an adjustable hyper-parameter in the range of 1 to N.
9. A method for few-view computed tomography (CT) image reconstruction, the method comprising:
applying, by a preprocessing module, a ramp filter to an input sinogram to yield a filtered sinogram;
receiving, by a first generator network, the filtered sinogram;
learning, by the first generator network, a filtered back-projection operation;
providing, by the first generator network, a first reconstructed image as an output, the first reconstructed image corresponding to the input sinogram; and
determining, by a discriminator network, whether a received image corresponds to the first reconstructed image or a corresponding ground truth image, the first generator network and the discriminator network corresponding to a Wasserstein generative adversarial network (WGAN), the WGAN optimized using an objective function based, at least in part, on a Wasserstein distance and based, at least in part, on a gradient penalty.
10. The method of claim 9 , further comprising:
receiving, by a second generator network, a concatenation of the first reconstructed image and a filtered back-projection of the input sinogram;
providing, by the second generator network, a second reconstructed image; and
determining, by the discriminator network, whether the received image corresponds to the second reconstructed image.
11. The method of claim 10 , further comprising receiving, by a filtered back projection module, the input sinogram and providing, by the filtered back projection module, the filtered back-projection of the input sinogram.
12. The method of claim 9 , wherein the first generator network is configured to learn the filtered back-projection operation in a point-wise manner.
13. The method of claim 9 , wherein the first generator network comprises a filtration portion, a back-projection portion, and a refinement portion.
14. The method of claim 9 , further comprising learning, by the first generator network, an initial filtered back-projection operation using image data from an image database comprising a plurality of images.
15. The method of claim 9 , wherein the first generator network is configured to reconstruct the first reconstructed image using O(C×N×N,) parameters, where N is a dimension of the first reconstructed image, N, is a number of projections, and C is an adjustable hyper-parameter in the range of 1 to N.
16. A computer readable storage device having stored thereon instructions configured for a few-view computed tomography (CT) image reconstruction, the instructions that when executed by one or more processors result in the following operations comprising:
applying a ramp filter to an input sinogram to yield a filtered sinogram;
receiving the filtered sinogram;
learning a filtered back-projection operation;
providing a first reconstructed image as an output, the first reconstructed image corresponding to the input sinogram; and
determining whether a received image corresponds to the first reconstructed image or a corresponding ground truth image, the operations corresponding to a Wasserstein generative adversarial network (WGAN), the WGAN optimized using an objective function based, at least in part, on a Wasserstein distance and based, at least in part, on a gradient penalty.
17. The device of claim 16 , wherein the instructions that when executed by the one or more processors result in the following additional operations comprising:
receiving a concatenation of the first reconstructed image and a filtered back-projection of the input sinogram;
providing a second reconstructed image; and
determining whether the received image corresponds to the second reconstructed image.
18. The device of claim 16 , wherein the filtered back-projection operation is learned in a point-wise manner.
19. The device of claim 16 , wherein the instructions that when executed by the one or more processors result in the following additional operations comprising learning an initial filtered back-projection operation using image data from an image database comprising a plurality of images.
20. The device of claim 16 , wherein the first reconstructed image is reconstructed using O(C×N×N,) parameters, where N is a dimension of the first reconstructed image, N, is a number of projections, and C is an adjustable hyper-parameter in the range of 1 to N.