Methods of spatial normalization of positron emission tomography images
An adaptive template image for registering a PET or a SPECT image includes a template image model including variability of values for each voxel in a template image according to one or more control parameters.
1. An adaptive template image system configured to perform spatial normalization and registration of a PET or a SPECT image using a single image registration procedure, the system comprising:
a storage configured to store and provide an adaptive template image comprising a template image model including a variability of values for one or more voxels in the adaptive template image, wherein the variability of values for each voxel ranges from a value corresponding to a normal level of uptake of an imaging agent to an abnormal level of uptake of the imaging agent, and wherein the imaging agent includes a dopamine transport agent DaTSCAN and a level of uptake is determined for a striatum, the variability of values adjustable using a control parameter x,
wherein the values of the voxels are defined by an equation:
I i template =I i 0 +I i slope x,
and wherein I i template is the value of the voxel i in the adaptive template image, I i 0 represents a value corresponding to a normal level of uptake of the imaging agent within that voxel, I i slope corresponds to a maximum additional amount of uptake of the imaging agent corresponding to a high level uptake of the imaging agent in that voxel with respect to the control parameter x, wherein the adaptive template image is in a reference space and is used for registering the PET or the SPECT image in the reference space during the single image registration procedure; and
an image processor configured to at least:
initiate a first registration of the PET or SPECT image into the reference space;
spatially normalize the PET or SPECT image to the adaptive template image in the first registration by:
(i) adjusting the template image model using a first control parameter, the adjusting to include adjusting among the variability of values for one or more voxels in the adaptive template image based on the first control parameter,
(ii) rendering the adaptive template image from the template image model,
(iii) performing a spatial normalization operation by adjusting the PET or SPECT image using spatial transformation parameters,
(iv) evaluating the adaptive template image and the PET or SPECT imaging to determine a convergence of the adaptive template image to the PET or SPECT image, wherein the convergence is determined by comparing voxels of the adaptive template image to corresponding voxels of the PET or SPECT image, and wherein image registration between the adaptive template image and the PET or SPECT image is not performed until the adaptive template image converges to fit the PET or SPECT image, and
(v) when the adaptive template image does not converge to fit the PET or SPECT image, continuously and adaptively altering both the adaptive template image and the PET or SPECT image by selecting an updated control parameter to render the adaptive template image and updated spatial transformation parameters to spatially normalize the PET or SPECT image for re-evaluation at (iv) within the single registration procedure and without performing the image registration until the adaptive template image converges to fit the PET or SPECT image; and
when the adaptive template image converges to fit the PET or SPECT image, register the PET or SPECT image in the reference space to complete the first registration.
2. The adaptive template image system of claim 1 , wherein the level of uptake is measured in gray matter of a brain.
3. The adaptive template image system of claim 1 , wherein the variability of values for each voxel ranges from a value corresponding to a normal level of amyloid in grey matter to a raised amyloid level in grey matter.
4. The adaptive template image system of claim 1 , wherein the control parameter is selected as part of a spatial normalization optimization process.
5. The adaptive template image system of claim 1 , wherein I 0 and I slope are determined using regression analysis of multiple images spanning different levels of disease state.
6. A method of performing spatial normalization of a PET or a SPECT image, using a single image registration procedure, and of registering the PET or the SPECT image to an adaptive template image comprising a template image model comprising variability of values for one or more voxels in the adaptive template image, wherein the variability of values for each voxel ranges from a value corresponding to a normal level of uptake of an imaging agent to an abnormal level of uptake of the imaging agent, and wherein the imaging agent includes a dopamine transport agent DaTSCAN and a level of uptake is determined for a striatum, the variability of values adjustable using a control parameter x, wherein the values of the voxels are defined by an equation:
I i template =I i 0 +I i slope x,
and wherein I i template is the value of the voxel i in the adaptive template image, I i 0 represents a value corresponding to a normal level of uptake of the imaging agent within that voxel, I i slope corresponds to a maximum additional amount of uptake of the imaging agent corresponding to a high level uptake of the imaging agent in that voxel with respect to the control parameter x,
wherein the adaptive template image is in a reference space and is used for registering the PET or the SPECT image in the reference space during the single image registration procedure, the method comprising:
initiating a first registration of the PET or SPECT image into the reference space;
spatially normalizing the PET or SPECT image to the adaptive template image in the first registration by:
adjusting the template image model using a first control parameter, the adjusting to include adjusting among the variability of values for one or more voxels in the adaptive template image based on the first control parameter;
rendering the adaptive template image from the template image model;
performing a spatial normalization operation by adjusting the PET or SPECT image using spatial transformation parameters;
evaluating the adaptive template image and the PET or SPECT imaging to determine a convergence of the adaptive template image to the PET or SPECT image, wherein the convergence is determined by comparing voxels of the adaptive template image to corresponding voxels of the PET or SPECT image, and wherein image registration between the adaptive template image and the PET or SPECT image is not performed until the adaptive template image converges to fit the PET or SPECT image, and
when the adaptive template image does not converge to fit the PET or SPECT image, continuously and adaptively altering both the adaptive template image and the PET or SPECT image by selecting an updated control parameter to render the adaptive template image and updated spatial transformation parameters to spatially normalize the PET or SPECT image for re-evaluation within the single registration procedure and without performing the image registration until the adaptive template image converges to fit the PET or SPECT image; and
when the adaptive template image converges to fit the PET or SPECT image, registering the PET or SPECT image in the reference space to complete the first registration.
7. The method of claim 6 , wherein said spatial normalization further comprises performing linear and non-linear transformations on the PET or SPECT image to find a match between the template image model and the PET or SPECT image, and wherein said linear transformation comprising at least one of translation, rotation, and scaling.
8. A system for performing spatial normalization of a PET or a SPECT image using a single image registration procedure, and for registering the PET or the SPECT image to an adaptive template image comprising a template image model including a variability of values for one or more voxels in the adaptive template image, wherein the variability of values for each voxel ranges from a value corresponding to a normal level of uptake of an imaging agent to an abnormal level of uptake of the imaging agent, and wherein the imaging agent includes a dopamine transport agent DaTSCAN and a level of uptake is determined for a striatum, the variability of values adjustable using a control parameter x, wherein the values of the voxels are defined by an equation:
I i template =I i 0 +I i slope x,
and wherein I i template is the value of the voxel i in the adaptive template image, I i 0 represents a value corresponding to a normal level of uptake of the imaging agent within that voxel, I i slope corresponds to a maximum additional amount of uptake of the imaging agent corresponding to a high level uptake of the imaging agent in that voxel with respect to the control parameter x,
wherein the adaptive template image is in a reference space and is used for registering the PET or the SPECT image in the reference space during the single image registration procedure, the system configured to:
during a first registration of the PET or SPECT image into the reference space, spatially normalize the PET or SPECT image to the adaptive template image in the first registration by:
adjusting the template image model using a first control parameter, the adjusting to include adjusting among the variability of values for one or more voxels in the adaptive template image based on the first control parameter,
rendering the adaptive template image from the template image model,
performing a spatial normalization operation by adjusting the PET or SPECT image using spatial transformation parameters,
evaluating the adaptive template image and the PET or SPECT imaging to determine a convergence of the adaptive template image to the PET or SPECT image, wherein the convergence is determined by comparing voxels of the adaptive template image to corresponding voxels of the PET or SPECT image, and wherein image registration between the adaptive template image and the PET or SPECT image is not performed until the adaptive template image converges to fit the PET or SPECT image, and
when the adaptive template image does not converge to fit the PET or SPECT image, continuously and adaptively altering both the adaptive template image and the PET or SPECT image by selecting an updated control parameter to render the adaptive template image and updated spatial transformation parameters to spatially normalize the PET or SPECT image for re-evaluation within the single registration procedure and without performing the image registration until the adaptive template image converges to fit the PET or SPECT image; and
when the adaptive template image converges to fit the PET or SPECT image, register the PET or SPECT image in the reference space to complete the first registration.
9. A non-transitory computer readable storage medium comprising computer readable program code including instructions for performing spatial normalization of a PET or a SPECT image, using a single image registration procedure, and for registering the PET or the SPECT image to an adaptive template image comprising a template image model comprising variability of values for one or more voxels in the adaptive template image, wherein the variability of values for each voxel ranges from a value corresponding to a normal level of uptake of an imaging agent to an abnormal level of uptake of the imaging agent, and wherein the imaging agent includes a dopamine transport agent DaTSCAN and a level of uptake is determined for a striatum, the variability of values adjustable using a control parameter x, wherein the values of the voxels are defined by an equation:
I i template =I i 0 +I i slope x,
and wherein I i template is the value of the voxel i in the adaptive template image, I i 0 represents a value corresponding to a normal level of uptake of the imaging agent within that voxel, I i slope corresponds to a maximum additional amount of uptake of the imaging agent corresponding to a high level uptake of the imaging agent in that voxel with respect to the control parameter x,
wherein the adaptive template image is in a reference space and is used for registering the PET or the SPECT image in the reference space during the single image registration procedure, wherein execution of the computer readable program code causes a processor to carry out a method comprising:
initiating a first registration of the PET or SPECT image into the reference space;
spatially normalizing the PET or SPECT image to the adaptive template image in the first registration by:
adjusting the template image model using a first control parameter, the adjusting to include adjusting among the variability of values for one or more voxels in the adaptive template image based on the first control parameter;
rendering the adaptive template image from the template image model;
performing a spatial normalization operation by adjusting the PET or SPECT image using spatial transformation parameters;
evaluating the adaptive template image and the PET or SPECT imaging to determine a convergence of the adaptive template image to the PET or SPECT image, wherein the convergence is determined by comparing voxels of the adaptive template image to corresponding voxels of the PET or SPECT image, and wherein image registration between the adaptive template image and the PET or SPECT image is not performed until the adaptive template image converges to fit the PET or SPECT image, and
when the adaptive template image does not converge to fit the PET or SPECT image, continuously and adaptively altering both the adaptive template image and the PET or SPECT image by selecting an updated control parameter to render the adaptive template image and updated spatial transformation parameters to spatially normalize the PET or SPECT image for re-evaluation within the single registration procedure and without performing the image registration until the adaptive template image converges to fit the PET or SPECT image; and
when the adaptive template image converges to fit the PET or SPECT image, registering the PET or SPECT image in the reference space to complete the first registration.
10. A method of constructing an adaptive template image comprising a template image model including a variability of values for one or more voxels in the adaptive template image, the method comprising:
transforming a set of PET or SPECT images into a reference space using spatial transformation parameters;
calculating a regression model for each voxel in the set of PET or SPECT images having been transformed into the reference space, where a dependent variable represents voxel intensities in the set of PET or SPECT images and is regressed on some variable which for each input PET or SPECT image represents a “true” value of where on a scale from a normal level of uptake to a high level of uptake of an imaging agent the voxel is located; and
generating, based on the template image model and a control parameter, the adaptive template image for registering, in a single registration procedure, the PET or SPECT image,
wherein the PET or SPECT image is to be registered by at least:
spatially normalizing, in a first registration of the PET or SPECT image into the reference space, the PET or SPECT image to the adaptive template image in the first registration by:
adjusting the template image model using a first control parameter x, the adjusting to include adjusting among the variability of values for one or more voxels in the adaptive template image based on the first control parameter x,
wherein the variability of values for each voxel ranges from a value corresponding to a normal level of uptake of an imaging agent to an abnormal level of uptake of the imaging agent, and wherein the imaging agent includes amyloid imaging agent and a level of uptake is determined for a striatum, wherein the values of the voxels are defined by an equation:
I i template =I i 0 +I i slope x,
and wherein I i template is the value of the voxel i in the adaptive template image, I i 0 represents a value corresponding to a normal level of uptake of the imaging agent within that voxel, I i slope corresponds to a maximum additional amount of uptake of the imaging agent corresponding to a high level uptake of the imaging agent in that voxel with respect to the first control parameter x,
rendering the adaptive template image from the adjusted template image model,
performing a spatial normalization operation by adjusting the PET or SPECT image using spatial transformation parameters,
evaluating the adaptive template image and the PET or SPECT image to determine a convergence of the adaptive template image to the PET or SPECT image, wherein the convergence is determined by comparing voxels of the adaptive template image to corresponding voxels of the PET or SPECT image, and wherein image registration between the adaptive template image and the PET or SPECT image is not performed until the adaptive template image converges to fit the PET or SPECT image, and
when the adaptive template image does not converge to fit the PET or SPECT image, continuously and adaptively altering both the adaptive template image and the PET or SPECT image by selecting an updated control parameter to render the adaptive template image and updated spatial transformation parameters to spatially normalize the PET or SPECT image for re-evaluation within the single registration procedure and without performing the image registration until the adaptive template image converges to fit the PET or SPECT image; and
when the adaptive template image converges to fit the PET or SPECT image, registering the PET or SPECT image in the reference space to complete the first registration.
11. The method of claim 10 , wherein the normal level of uptake corresponds to Aß− and the high level of uptake corresponds to Aß+.
12. The method of claim 10 , wherein the set of PET or SPECT images have been transformed into the reference space with a co-registered anatomical image.
13. A positron emission tomography (PET) system comprising:
a storage device configured to store and provide an adaptive template image comprising a template image model including a variability of values for one or more voxels in the adaptive template image, wherein the variability of values for each voxel ranges from a value corresponding to a normal level of uptake of an imaging agent to an abnormal level of uptake of the imaging agent, and wherein the imaging agent includes a dopamine transport agent DaTSCAN and a level of uptake is determined for a striatum, the variability of values adjustable using a control parameter x,
wherein the values of the voxels are defined by an equation:
I i template =I i 0 +I i slope x,
and wherein I i template is the value of the voxel i in the adaptive template image, I i 0 represents a value corresponding to a normal level of uptake of the imaging agent within that voxel, I i slope corresponds to a maximum additional amount of uptake of the imaging agent corresponding to a high level uptake of the imaging agent in that voxel with respect to the control parameter x,
wherein the adaptive template image is in a reference space and is used for registering a PET or a SPECT image in the reference space during single image registration procedure;
a detector for detecting positron emissions from a brain of a subject, wherein the detector generates signals representing the positron emissions that are stored in the storage device;
an image processor configured to at least:
initiate a first registration of the PET or SPECT image into the reference space;
spatially normalize the PET or SPECT image to the adaptive template image in the first registration by:
(i) adjusting the template image model using a first control parameter, the adjusting to include adjusting among the variability of values for one or more voxels in the adaptive template image based on the first control parameter,
(ii) rendering the adaptive template image from the template image model,
(iii) performing a spatial normalization operation by adjusting the PET or SPECT image using spatial transformation parameters,
(iv) evaluating the adaptive template image and the PET or SPECT imaging to determine a convergence of the adaptive template image to the PET or SPECT image, wherein the convergence is determined by comparing voxels of the adaptive template image to corresponding voxels of the PET or SPECT image, and wherein image registration between the adaptive template image and the PET or SPECT image is not performed until the adaptive template image converges to fit the PET or SPECT image, and
(v) when the adaptive template image does not converge to fit the PET or SPECT image, continuously and adaptively altering both the adaptive template image and the PET or SPECT image by selecting an updated control parameter to render the adaptive template image and updated spatial transformation parameters to spatially normalize the PET or SPECT image for re-evaluation at (iv) within the single registration procedure and without performing the image registration until the adaptive template image converges to fit the PET or SPECT image; and
when the adaptive template image converges to fit the PET or SPECT image, register the PET or SPECT image in the reference space to complete the first registration.
14. A positron emission tomography (PET) system comprising:
an image processor for performing spatial normalization and registration of a PET or a SPECT image, using a single image registration procedure, and for registering the PET the SPECT image to an adaptive template image comprising a template image model including a variability of values for one or more voxels in the adaptive template image, wherein the variability of values for each voxel ranges from a value corresponding to a normal level of uptake of an imaging agent to an abnormal level of uptake of the imaging agent, and wherein the imaging agent includes a dopamine transport agent DaTSCAN and a level of uptake is determined for a striatum, the variability of values adjustable using a control parameter x, wherein the values of the voxels are defined by an equation:
I i template =I i 0 +I i slope x,
and wherein I i template is the value of the voxel i in the adaptive template image, I i 0 represents a value corresponding to a normal level of uptake of the imaging agent within that voxel, I i slope corresponds to a maximum additional amount of uptake of the imaging agent corresponding to a high level uptake of the imaging agent in that voxel with respect to the control parameter x,
wherein the adaptive template image is in a reference space and is used for registering the PET or the SPECT image in the reference space during the single image registration procedure, the image processor configured to:
initiate a first registration of the PET or SPECT image into the reference space;
spatially normalize the PET or SPECT image to the adaptive template image in the first registration by:
adjusting the template image model using a first control parameter, the adjusting to include adjusting among the variability of values for one or more voxels in the adaptive template image based on the first control parameter;
rendering the adaptive template image from the template image model;
performing a spatial normalization operation by adjusting the PET or SPECT image using spatial transformation parameters;
evaluating the adaptive template image and the PET or SPECT imaging to determine a convergence of the adaptive template image to the PET or SPECT image, wherein the convergence is determined by comparing voxels of the adaptive template image to corresponding voxels of the PET or SPECT image, and wherein image registration between the adaptive template image and the PET or SPECT image is not performed until the adaptive template image converges to fit the PET or SPECT image, and
when the adaptive template image does not converge to fit the PET or SPECT image, continuously and adaptively altering both the adaptive template image and the PET or SPECT image by selecting an updated control parameter to render the adaptive template image and updated spatial transformation parameters to spatially normalize the PET or SPECT image for re-evaluation within the single registration procedure and without performing the image registration until the adaptive template image converges to fit the PET or SPECT image; and
when the adaptive template image converges to fit the PET or SPECT image, register the PET or SPECT image in the reference space to complete the first registration.
15. A computer-implemented method of performing spatial normalization of a PET or a SPECT image, using a single image registration procedure, and registering of the PET or the SPECT image to an adaptive template image comprising a template image model including a variability of values for one or more voxels in the adaptive template image, wherein the variability of values for each voxel ranges from a value corresponding to a normal level of uptake of an imaging agent to an abnormal level of uptake of the imaging agent, and wherein the imaging agent includes a dopamine transport agent DaTSCAN and a level of uptake is determined for a striatum, the variability of values adjustable using a control parameter x,
wherein the values of the voxels are defined by an equation:
I i template =I i 0 +I i slope x,
and wherein I i template is the value of the voxel i in the adaptive template image, I i 0 represents a value corresponding to a normal level of uptake of the imaging agent within that voxel, I i slope corresponds to a maximum additional amount of uptake of the imaging agent corresponding to a high level uptake of the imaging agent in that voxel with respect to the control parameter x,
wherein the adaptive template image is in a reference space and is used for registering the PET or the SPECT image in the reference space during the single image registration procedure, the method comprising:
initiating a first registration of the PET or SPECT image into the reference space;
spatially normalizing the PET or SPECT image to the adaptive template image in the first registration by:
adjusting the template image model using a first control parameter, the adjusting to include adjusting among the variability of values for one or more voxels in the adaptive template image based on the first control parameter;
rendering the adaptive template image from the template image model;
performing a spatial normalization operation by adjusting the PET or SPECT image using spatial transformation parameters;
evaluating the adaptive template image and the PET or SPECT imaging to determine a convergence of the adaptive template image to the PET or SPECT image, wherein the convergence is determined by comparing voxels of the adaptive template image to corresponding voxels of the PET or SPECT image, and wherein image registration between the adaptive template image and the PET or SPECT image is not performed until the adaptive template image converges to fit the PET or SPECT image, and
when the adaptive template image does not converge to fit the PET or SPECT image, continuously and adaptively altering both the adaptive template image and the PET or SPECT image by selecting an updated control parameter to render the template adaptive image and updated spatial transformation parameters to spatially normalize the PET or SPECT image for re-evaluation within the single registration procedure and without performing the image registration until the adaptive template image converges to fit the PET or SPECT image; and
when the adaptive template image converges to fit the PET or SPECT image, registering the PET or SPECT image in the reference space to complete the first registration.