6 results about "Correction for attenuation" patented technology

The invention is suitable for the technical field of medical images, and provides a correction method and device for a single photon emission type computed tomography image, electronic equipment and a storage medium, and the method comprises the steps: obtaining a training sample set which comprises a plurality of training samples, each training sample comprises an uncorrected SPECT image, a CT image registered with the uncorrected SPECT image, an attenuation image obtained by converting the CT image, and a corrected SPECT image obtained by performing attenuation correction on the uncorrected SPECT image by using the attenuation image; the training sample set is used for training a first correction network and a first generation network, the first correction network is used for generating a corrected SPECT image according to the input uncorrected SPECT image, and the second generation network is used for generating a target image according to the input corrected SPECT image; and performing iterative cascade fine tuning on the trained first correction network and the first generation network by using the training sample set to obtain a second correction network and a second generation network.

A system and method includes determining a region of interest of an imaged object, generating a first linear attenuation coefficient map of the imaged object, the first linear attenuation coefficient map being generated to associate voxels of the region of interest of the imaged object with a greater linear attenuation coefficient than voxels of other regions of the imaged object, attenuation correcting a plurality of tomographic frames of the imaged object based on the first linear attenuation coefficient map to generate a second plurality of tomographic frames, and determining a tomographic inconsistency of the second plurality of tomographic frames. Some aspects further include generating a second linear attenuation coefficient map of the imaged object, attenuation correcting the plurality of tomographic frames based on the second linear attenuation coefficient map to generate a third plurality of tomographic frames, and reconstructing a three-dimensional image based on the third plurality of tomographic frames and the determined tomographic inconsistency.

Systems and methods for generating registered attenuation maps are disclosed. For example, positron emission tomography (PET) measurement data, and co-modality measurement data from an anatomy modality, such as computed tomography (CT) data, is received from an image scanning system. A histo-image is generated based on the PET measurement data, and a co-modality image is generated based on the co-modality measurement data. A trained machine learning process is applied to the histo-image and the co-modality image. The trained machine learning process is configured to correct for misalignment between the histo-image and the co-modality image. Based on the application of the trained machine learning process to the histo-image and the co-modality image, a registered attenuation map is generated. In some examples, a PET image is reconstructed using the registered attenuation map.

The invention relates to the technical field of nuclear medicine imaging, and particularly provides an SPECT image reconstruction method and device applying scattered photons and a medium, and the method comprises the steps: determining an energy window group according to radionuclides, and obtaining an attenuation correction image and a DRR image based on a CT image; performing data acquisition based on the energy window group; inputting the DRR image and all the energy window data into a projection domain scatter correction model to enable the projection domain scatter correction model to output low-noise projection data without scattering photons, and then performing tomographic reconstruction on the low-noise projection data to generate a low-noise tomographic reconstruction image; or performing tomographic reconstruction on each piece of energy window data to obtain a tomographic reconstruction image corresponding to each piece of energy window data, and then inputting the attenuation correction image and all the tomographic reconstruction images into an image domain scatter correction model to enable the image domain scatter correction model to output a low-noise tomographic reconstruction image; the method can effectively improve the application value of the SPECT technology in precision medical treatment and quantitative analysis.

A system based upon artificial neural networks generates attenuation-corrected SPECT from non-attenuation-corrected SPECT (single photon emission computed tomography) without or with an intermediate step of attenuation map estimation. The system includes a SPECT scanner with CZT cameras for dynamic SPECT imaging. The system also includes a machine learning system including a 3D Dual Squeeze-and-Excitation Residual Dense Network for generating attenuation-corrected SPECT or attenuation maps from non-attenuation-corrected SPECT. The machine learning system reconstructs images from photopeak window and one or more scatter windows of the SPECT scanner are fed to the 3D Dual Squeeze-and-Excitation Residual Dense Network to generate attenuation-corrected SPECT or attenuation maps.

A system based upon artificial neural networks generates attenuation-corrected SPECT from non-attenuation-corrected SPECT (single photon emission computed tomography) without or with an intermediate step of attenuation map estimation. The system includes a SPECT scanner with CZT cameras for dynamic SPECT imaging. The system also includes a machine learning system including a 3D Dual Squeeze-and-Excitation Residual Dense Network for generating attenuation-corrected SPECT or attenuation maps from non-attenuation-corrected SPECT. The machine learning system reconstructs images from photopeak window and one or more scatter windows of the SPECT scanner are fed to the 3D Dual Squeeze-and-Excitation Residual Dense Network to generate attenuation-corrected SPECT or attenuation maps.